Recommendations???

[traynor]

I have read most of the posts on this forum, and I am more confused than I was before I started. I have a modest amount of expertise using computer applications, including programming. I am familiar with Sartin's old applications, and most of the newer ones. The problem is that I have never found an application that could do more than I can do with a graphing calculator and a (very) simplistic pace formula.

Any recommendations as to software applications you are using, or have used, that actually do more than massage a few numbers would be greatly appreciated. I am not a big fan of subscription downloads, primarily because I have not found an application that can generate a positive return large enough to warrant the cost.

If you had to recommend a software application to someone with the primary intention of using it to generate a profit (rather than cash a lot of tickets, a whole separate issue), what would it be?

Has anyone used Kaywood's application with any degree of success? Or the Trackmaster application? (again, over time; it seems like anything will win for a short period or in a few races at some track somewhere).

Last question; IS THERE any application out there that can actually generate a profit on flat win bets over time?

If anyone is working on an application of his or her own, and would like to trade information, ideas, insights, or wild conjectures, I have a reasonable level of skill with both Java and Visual Basic and my primary interest is in developing an application for personal use (unless someone else has already developed one that can crank out a positive expectation). Collaboration and multiple viewpoints are always useful.
Thanks

[andicap]

All depends on what kind of handicapper you are and what kind of tool you're looking for.
No one here will recommend a "black box" which will pick winners. That's chasing Fool's Gold.
Are you looking for something to do pace analysis?
Are you looking for something that will get you overlaid long-shot horses with low win % but huge payoffs?
Are you looking for a complicated database program that does a billion thiings but you've got to decide the handful you want to focus on?

If you don't know your identity as a handicapper a computer program will be a waste of money and do you more harm than good.

I speak from experience. In the early 90s I began going through a ton of them thinking they would me winners, but lacked a solid foundation in pace or a set of core handicapping beliefs (In the 80s I had used Beyer figs, trips, biases but with no simulcasting played only 1 track.). The results were disastrous as I stumbled from one highly hyped program to another.

Finally today I know what I'm looking to do in a race and the program I use is merely a tool to crunch numbers and point me to horses who might run well at overlaid odds according to the criteria I look for. I'm a pace-form cycle player and for that HTR fills the bill. But it wouldn't help me if I didn't know how to interpret and analyze its output.

[Blackgold]

Tray, there is no black box.

Since there are so many different type of races. . mdn, cl, alw, hcp, etc. . .no one software will pick 'da winner all the time.

In fact, over time, most software will produce a loss equal to the track take., i.e. RANDOM.

My suggesstion is to gravitate toward something simple. Simple ratings that can be used in context with your skills to evaluate whether there is profit opportunity or not.

Some do very well with the Sheets. I tried them during that free period during the end of the year. Maybe with more practice I could become more proficient with them.

Tried ALLWAYS some years back, but my impression is that it's a program cramed with all the best ideas of many good people and it sometimes clouds the picture rather than presents a clear picture.

Some construct their own numbers and do so with the goal of uncovering overlays. It is overlays that will give you a consistent profit over time.

I've made consistent money with Handicapping Magic. I like the simple numbers.

When I see a standout runner that figures to go off at high odds, I then try and take apart the race and box with others so if my runner comes in, I get paid.

To pick apart a race you need all your skills. Every book you've read on class, speed, pace, trainer moves, developing young horses, tote analysis and even body language. And with all that, you still get beat.

Case in point. I liked the 9 yesterday in the Risen Star at FG. Was all over the runner in the win hole and in the exotics with 6 other runners and still did not cash when he ran 2nd at 40-1.

My Italian buddy was playing poker in AC and darting back and forth to the race book to send in the races we collaborated on. Two minutes before post of the Risen Star his wife calls and says "bet the 8 for me to win in the Risen Star 'cause I'm watching TVG and the 8 looks good in warmups." The 8 won at 22-1 and the exacta with my selection paid somebody, not me, $369 for a buck. Go figure.

I can console myself with the fact that I was fishing in the right pond and there was a whopper there, it just ended up on someone elses hook.

Find a rating system you are comfortable with and learn everything you can about it and then watch, and wait for opportunities and then cast your net. If you are going after overlays, then over time you will prevail.

If you are looking for a blackbox, then others will prevail over you.

[speedking]

Blackgold,

I can sympathize with you. I didn't feel that the 9 horse in the Risen Star was necessarily the best, but he was certainly a tremendous value. I keyed him back and forth with 4 others in exactas and tris, even hit the all button for third on one tri ticket. Thought about it for a minute and decided to cover myself with a last second win bet. Never used the 8 of course. Good call, but nothing to show for it except a lower account balance.

speedking

[betovernetcapper]

I think most good software programs will be effective at some tracks-distances-surfaces some of the time. There seems to be a sort of ebb and flow. What I attempt to do with Netcapper is to monitor which tracks-distances it's working at and play them and attempt to drop the track when it becomes cold. This is not always an easy thing to do.

[AQUEBUCKS]

I had the (9) (small VTR with odds of 5-1) also, but could not come up with the (8) anywhere. I played all kinds of exactas/tripples to no avail. When I saw the place price I just cringed. I feel your pain, SW3=n

Aquebucks:confused:

[TonyK@HSH]

Traynor,

I know of several winning players using HSH (HorseStreet). I'd suggest you give Dave Schwartz a call to see if it blends with your personal style of play.

Good luck

Tony

[hdcper]

Traynor,

I would suggest you try some of the high quality programs with demos first, such as, Jcapper, Equisim and HTR. All three have super programmers that provide great customer service, a strong product and the opportunity to try it before you buy it.

Further, check out their websites and ask your questions:

http://jcapper.com/

http://thorotech.com/

http://www.homebased2.com/km/msiez9987.htm

Hope you find what your looking for,

Hdcper

[NoDayJob]

I have a modest amount of expertise using computer applications, including programming. The problem is that I have never found an application that could do more than I can do with a graphing calculator and a (very) simplistic pace formula.

:) Welcome---

I believe you've answered your own question.


Last question; IS THERE any application out there that can actually generate a profit on flat win bets over time?

Out of the box, probably not. At least not without extensive research and modification on your part. IMO, with your programming backround, you'd do much better to write your own software.

NDJ

[46zilzal]

Mutually exclusive disciplines...MOST programs help in the former, FEW in the latter..just have to look with different eyes..SO MANY people KEEP CHANGING programs looking for the HOLY GRAIL without EVER really LEARNING the program they have: You NEED to stay with one at least a month to know what you have.

KEEP A JOURNAL of both 'capping and wagering decisions to find out YOUR weaknesses before you jump...IF what you have works...DON'T CHANGE IT

[JimG]

.SO MANY people KEEP CHANGING programs looking for the HOLY GRAIL without EVER really LEARNING the program they have: You NEED to stay with one at least a month to know what you have.



Hi there,

I stongly agree that program jumpers (like me) many times do not learn the program completely before moving on to others. I have certainly been guilty of that in the past! I would say though that if you purchase a full featured software, you should spend a minimum of 6 months with it learning as much as you can about it. Good software takes a strong committment from the user to reach maximum effectiveness.

Jim

PS...I've lost money over the years jumping from software to software. To any newbies reading this, please do not repeat my mistakes.

[kitts]

I have done my share of jumping. I now use only All-In-One and it fits me well so that I can make some money using it. Software is a tough choice but don't lose sight of the fact that any software is only a tool, not a solution. If the software "fits" then that's the one to use.

[fast4522]

Check out new Sartin or any Sartin here:
VDCMessengerService-subscribe@yahoogroups.com

[raybo]

I agree with most of what has been said here. IMO, software is only a tool to be used as an advantage over the public. IMO, publically available software will not produce a profit over time on flat win, place, or show bets.

In order to make profits with any flat bet system over time, one would have to become extremely familiar with the software, keep it updated (or "learning"), and be very picky about when you put your money at risk. I personally have produced a profit with my automated spreadsheet program but not on win, place and show bets. The payouts vs. win % just isn't there. There is, I'm sure, some excellent software available today, using artificial intellligence and the like that will provide a knowledgeable user with a distinct advantage over the public, but I doubt that any could be used strictly by themselves with no input from the user.

If you can program, no matter which language or application, and you know how to handicap "by hand" already, then it seems clear to me that you should find a way to automate your "by hand" methods and design your system to be easily updateable. You can do a pretty good job of analysis using Excel and it doesn't take a computer geek to write it. I personally use Brisnet's data in mine and it's a snap to download and input into the spreadsheet.

So, do it yourself.

Remember, half the fun is getting there!

[kingfin66]

Where have you been?

[raybo]

re: "Where have you been?"

?? Are you being facetious or do you really care that I haven't posted on this board for quite some time? The site your link sent me to holds nothing special concerning my previous remarks that I can see. I simply added my personal opinions concerning commercially available handicapping software to that of the others in the thread. If you have something specific in mind then, by all means, be specific. If you really did mean, "Why haven't you posted in a while", my appologies if this sounds defensive, but offensive posts are not unknown here so you can understand my stance concerning your statement.

[ratpack]

I also agree with most of what has been said here.

I think what you have to be concerned about first is "Do you fit the Program"

Some programs are great but have 17 pages of data you have to scroll through to make a selection then by the time the races roll around you are to burned out to make a clear betting decision.

Some people like Handicapper have great success using 2 different programs that compliment each other, me I got more confused using 2 programs.

My suggestion and as other have said take your time and check out as many demos and free trials as possible.

One other thing and this is just my opinion but these trial periods are usually not enough time to decide if a program is going to be profitable in the long run so I would use it more to see if it is going to be something you can be comfortable with using long term

[kingfin66]

Raybo,

I was being totally sincere. In the past, you posted useful info about how you play superfectas. I also recall, unfortunately, that you were subjected to some unkind comments. Please feel flattered that somebody would notice your absence. There was absolutely no sarcasm intended in my question.

As far as the link in my signature, it is for my brother's site. I am not trying to sell anything to anybody here, and you're right, it holds no relevance to any comments you made - it is a dynamic link (appears in all of my posts).

Your response in this thread is dead on. I was reading through the thread, saw your post, and said to myself, "damn, I haven't seen Raybo here in a long time."

Take care.

[cj]

If you can program, no matter which language or application, and you know how to handicap "by hand" already, then it seems clear to me that you should find a way to automate your "by hand" methods and design your system to be easily updateable.

I think you are simplifying things a bit. A horse race can be at least as complicated as a chess match. How many lines of code do you think there are in the Deep Blue program?

[raybo]

Kingfin66,

Well, my sincere appologies indeed, I am truly flattered that you found my previous posts interesting enough to remember who wrote them.

To answer your question, I just haven't visited this forum in quite some time, until last week. I generally don't do the forum thing unless I am seeking information or opinion or I have discovered something I feel may help someone else.

I had a profitable year last year ($4500 net, 39% profit, 7 IRS reportables, :( unfortunately), even after quite a dry spell in the fall. As a result of that extended losing streak, I found myself approaching a "tilt" and decided a break was in order. I have done very little regarding the ponies since October. I found myself doubting my own abilities as a handicapper and that didn't feel too good, to say the least. It's nothing new, happens to many in our field, but it was my first experience with that degree of self doubt. My 'capping partner and I began again the first of this month, so I decided to see what was going on in this forum. I saw this thread from, apparently, someone who had some similar questions concerning handicapping software that I had had several years ago, before I designed my spreadsheet. I thought that my opinion might help him decide in favor of "doing his own thing", which I feel is the most rewarding approach for a capable handicapper who can ask himself the right kind of questions and then convert those questions into a viable computerized analysis method.

Thank you for asking about my absence!

[raybo]

cj,

I'm not sure if I should be offended or flattered at being accused of "simplifying" something as complex, confusing, and stressful as thoroughbred handicapping. I wasn't suggesting that he write code that could solve the world's problems. He appeared to have an adequate knowledge of handicapping already and some computer knowledge as well, and he appeared to be looking for an application that would help him further his current capabilities. Maybe I misunderstood him. Maybe he wants to buy a program that will pick the winners and produce a hefty profit just by clicking a button. If you have one of those then, by all means, sell it to him, but please don't try to sell it to me.

[acorn54]

[QUOTE=raybo]Kingfin66,


I had a profitable year last year ($4500 net, 39% profit, 7 IRS reportables,

i think you are doing great for the year, 40% profit may be as good as it gets
acorn

[cj]

Whoa raybo, what's with the venom? All I was saying was that some things, which on the surface would appear to be easily programmable, are not. That's it. How that turned into me trying to sell you something is beyond me I guess.

[sjk]

CJ,

I don't buy the Deep Blue analogy. That is a program that aspires to beat every chess player in the world.

If a handicapping program can beat 99% of the players in the world it can probably do pretty well (if that is not the right number pick your own). It's possible that an individual who was willing to invest a reasonable amount of time and study could write a chess program that could beat 99% of the players in the world.

I don't plan to try since I don't play chess and as far as I can tell mediocre chess players are not lining up to send their money to the operator of such a program whereas mediocre race players do it every day.

[kingfin66]

Kingfin66,

Well, my sincere appologies indeed, I am truly flattered that you found my previous posts interesting enough to remember who wrote them.

To answer your question, I just haven't visited this forum in quite some time, until last week. I generally don't do the forum thing unless I am seeking information or opinion or I have discovered something I feel may help someone else.

I had a profitable year last year ($4500 net, 39% profit, 7 IRS reportables, :( unfortunately), even after quite a dry spell in the fall. As a result of that extended losing streak, I found myself approaching a "tilt" and decided a break was in order. I have done very little regarding the ponies since October. I found myself doubting my own abilities as a handicapper and that didn't feel too good, to say the least. It's nothing new, happens to many in our field, but it was my first experience with that degree of self doubt. My 'capping partner and I began again the first of this month, so I decided to see what was going on in this forum. I saw this thread from, apparently, someone who had some similar questions concerning handicapping software that I had had several years ago, before I designed my spreadsheet. I thought that my opinion might help him decide in favor of "doing his own thing", which I feel is the most rewarding approach for a capable handicapper who can ask himself the right kind of questions and then convert those questions into a viable computerized analysis method.

Thank you for asking about my absence!

You had an awesome year! As I recall, your ROI was extremely high (> 200%?) for a time earlier in the year. With your focus on the super, you're going to have some runouts. The way you handled the streak will make you a long-term consistent winner (like you already are no doubt). Feeling yourself going on tilt and stopping, or even just slowing down, as opposed to increasing bet size, or reducing selectivity in the name of getting your losses back takes tremendous discipline. I will often slow down my handicapping after the Breeder's Cup because there are so many distractions in November and December, plus it gives me a mental break.

Congrats again on a great year of playing the horses.

[raybo]

cj

Maybe I over-reacted, but when someone replies to a post negatively, who in their reply is advertizing a handicapping method, program, or tool , I naturally become defensive.

I stand by my original statement that someone with an adequate handicapping ability and the ability to automate, via computer, might be better off doing their own work, rather than shopping for an existing product that may or may not treat the "numbers" in a similar manner. One does not have to write a "Deep Blue" program for racing to attain something that is effective for their own purposes. All the existing programs I have seen will not do what I consider to be the most important factor in racing anyway, figure current condition. Really, all an application needs to do is crunch the numbers for you, you have to decide whether each horse can run those nuimbers today. Crunching numbers is not hard to do with just a little bit of computer savvy.

[cj]

Raybo,
You'll get no argument from me on what you say. I didn't understand that was what you meant about the programming, certainly computers are wonderful number crunchers.

[traynor]

Thank you to everyone who posted recommendations to answer my questions. For some obscure reason, I have been unable to post on this forum again until today. A little frustrating.

My understanding of pace is that it is highly subjective, much as Bill Quirin's response to Beyer's question about what he did with his pace numbers; Quirin quickly replied, "I look at them" (to everyone's great amusement). In short, a list of numbers is presented that may or may not be predictive (or even useful), based on many other factors.

The approach I favor is a synthesis of pace handicapping and trip handicapping, with serious doses of inspection and body language mixed in. As fascinating as the idea of sophisticated computer applications selecting winners may be, the bottom line is that profit is the goal (at least for me), and that is strongly correlated with public perceptions and the mutuels that result from the perception.

I would really like to find a computer app that could predict enough winners to turn a profit on the days I don't have time to fully handicap a set of races (most days). Unfortunately, the more responsibility I turn over to the app, the more my bottom line deteriorates.

After about 15 years of almost complete immersion, I have been unable to find a computer application that produces results (not just endless columns of "ratings" that combine everything from jockey age to the phase of the moon) that are superior to the numbers generated with a simplistic Phase III type routine in a TI-83plus graphing calculator, with one exception. That exception is based on MDSS (managerial decision support software) that is a component of an ERP (enterprise resource planning) software bundle. The problem is that it takes an immense investment of time and effort to generate information that is useful, and even then, it is dynamic. That is, like all other "expert systems," it requires a continual process of real time adjustment to reflect current reality, as opposed to historical data.

Is anyone using a similar application, and, if so, what kind of responses are you getting with it? For those of you who may not be familiar with the terms, the apps are widely used in decision making in business, especially larger enterprises. Several sample apps are available as free downloads or shareware, and some even plug in to Excel for modeling. A search on "decision support software" might open a whole new world of opportunity for any interested in full time handicapping. I wouldn't recommend DSS for casual users; it takes a LOT of time and effort to create meaningful, useful data.

That is, crunching historical data (even with "sophisticated statistical analysis software") is simplistic, and fairly easy. Excel and SPSS have more statistical analysis capabilities that most users are capable of using effectively. Creating a dynamic probability predictive model to forecast future results is much more complex, much more interesting, and, fortunately, much more profitable.
Thanks

[sjk]

I have had good success with Access. I don't find current reality to be that much different than past history. I have used the same program (with incremental improvement over time) for seven or eight years and have not seen any fall off in effectiveness or any need for drastic overhaul.

[Kreed]

yes, i agree with what you posted. let's define a method/program. CJ's
is NOT a method/program ... it does NOT stick its neck out & say ,,,, #5 is
the Winner & #4 is 2nd. it rates all races numerically. It is NOT a system,
although I'm sure some have incorporated CJ's numbers into one. And, the
Computrak much touted quadratic equation, 1st in DOS by Daddy Engineer --
Now, on the NET by SonOfDaddy, is also NOT a system. it doesn't predict outcomes. That's why its NOT scientific --- its not testable. This is not a
slight, but systems are infinitely more complex that data keepers. Also, I am not saying that CJ's numbers or Computrak's ratings are not useful, it's just very hard to say how well they do in the long term. PS: my company spends close
to 5mil a year on PREDICTION SYTEMS -- in one area or another. Initially,
it spent ~30Mil just getting itself going. What's the MOST a system seller can
make? Judging from U tightwads here, maybe like $45 dollars profit, whereas,
wall st firms think biggER. lol ... 2nd PS: both CJ figs & Computrak's COULD
be testable easily if ONLY they defined their selection criteria. It's always
EZ to say "I had that winner" but yeah, just say so BEFORE.

[traynor]

sjk wrote <I have had good success with Access. I don't find current reality to be that much different than past history>

I have found the complete opposite; past history tends to smooth with the size of the sample, and apples and oranges in the same basket are only good for fruit salad. Everything is contingent on how people bet, and that changes with the meet, the season, even the weather. Consider the effect on the mutuels at GP with the influx of snowbirds as an obvious example. Picking x% winners can show a profit one month and a loss the next, based on nothing more than fluctuations in the mutuel pools.

Access is a simple database app, and I agree that it is adequate for most record keeping--as long as the input is limited to basics. That is not quite the same as DSS, the basic advantage of which is a continual, real time evaluation of results according to current reality--quite different than results based on a few anomalies ($45 winners for example) that rarely occur again. That is, the best model in the world using history cannot compare with the output of a DSS using the same data in an expert system. Not opinion. Fact.
Thanks

[raybo]

Posted by Kreed: "yes, i agree with what you posted. let's define a method/program. CJ's is NOT a method/program ... it does NOT stick its neck out & say ,,,, #5 is the Winner & #4 is 2nd. it rates all races numerically. It is NOT a system, although I'm sure some have incorporated CJ's numbers into one. And, the Computrak much touted quadratic equation, 1st in DOS by Daddy Engineer -- Now, on the NET by SonOfDaddy, is also NOT a system. it doesn't predict outcomes. That's why its NOT scientific --- its not testable. This is not a slight, but systems are infinitely more complex that data keepers. Also, I am not saying that CJ's numbers or Computrak's ratings are not useful, it's just very hard to say how well they do in the long term. PS: my company spends close to 5mil a year on PREDICTION SYTEMS -- in one area or another. Initially, it spent ~30Mil just getting itself going. What's the MOST a system seller can make? Judging from U tightwads here, maybe like $45 dollars profit, whereas, wall st firms think biggER. lol ... 2nd PS: both CJ figs & Computrak's COULD be testable easily if ONLY they defined their selection criteria. It's always EZ to say "I had that winner" but yeah, just say so BEFORE."

Kreed,

You hit the nail on the head. Most commercially available handicapping applications are not "systems". They are just one step beyond the "data" suppliers like Brisnet, etc. They are simply massaging the data into composites, not that the new composites aren't viable, because some of them are, but they are still just "data".

There are handicapping systems that do indeed stick their necks out, but to my knowledge, none have ever produced a flat win bet profit over time without user input in the decision making process, so none of these systems would fit Traynor's requirements as stated. He asked if a system exists that would do that, on it's on, without user input.

That is one reason that I decided to create my own "system" which does rank the entries in projected finish order. Because I wager on superfectas, this projected finish order was essential. And, yes, in testing without human input, it did produce a sizeable profit, at the only track that I ran through it. However, as we all know, this sport is not static, it is in constant flux. I would not attempt to use the system as a black box with no input into the final wagering order. Although it might again produce a profit that way, I would wager against it doing that over time. The key for any "system" producing a profit over time appears to be making selections occasionally that make very little sense, that actually hit for huge payouts. Without those huge payouts the system will lose money over time. With flat win bet wagering methods, to make a profit, even with a very good win %, you must place sizeable wagers at reasonable and higher odds. Even so, I'm not sure the meager reward would be worth the substantial risk involved. Commercial systems are going to select the highest graded horses, and most of us know, that is not where the profit lies.

[thoroughbred]

yes, i agree with what you posted. let's define a method/program. CJ's
is NOT a method/program ... it does NOT stick its neck out & say ,,,, #5 is
the Winner & #4 is 2nd. it rates all races numerically. It is NOT a system,
although I'm sure some have incorporated CJ's numbers into one. And, the
Computrak much touted quadratic equation, 1st in DOS by Daddy Engineer --
Now, on the NET by SonOfDaddy, is also NOT a system. it doesn't predict outcomes. That's why its NOT scientific --- its not testable. This is not a
slight, but systems are infinitely more complex that data keepers. Also, I am not saying that CJ's numbers or Computrak's ratings are not useful, it's just very hard to say how well they do in the long term. PS: my company spends close
to 5mil a year on PREDICTION SYTEMS -- in one area or another. Initially,
it spent ~30Mil just getting itself going. What's the MOST a system seller can
make? Judging from U tightwads here, maybe like $45 dollars profit, whereas,
wall st firms think biggER. lol ... 2nd PS: both CJ figs & Computrak's COULD
be testable easily if ONLY they defined their selection criteria. It's always
EZ to say "I had that winner" but yeah, just say so BEFORE.

KREED,
As you probably know, the latest CompuTrak version has an Oddsline and Summary report, and as far as prediction goes, the instructions tell you that the most straight forward way to predict the outcome is to go with the horse with the lowest odds of the Oddsline.

In general I would agree with you that posting results AFTER the race is not a good test. BUT, when I posted those great Aqueduct results because a customer called my attention to it. notice that I said anyone who has CompuTrak could check. It matters not that it was after the race, since the way to pick the horse was, as suggested, pick the one with the lowest odds of the Oddsline. In other words, the rule for picking the horses is part of CompuTrak instructions. So, in this case, looking at results AFTER the race is valid.

Of course, CompuTrak has additonal techniques for more extensive handicapping, but the demo on the web site: www.revelationprofits.com and the tutorial, etc. emphasize that the quick and easy way is to go with the horses with the lower odds as generated by the program.

[sjk]

Traynor,

I guess I would disagree with your "fact". My opinion is that the program that you use is less important than how you use the program. In my case the Access program does much more than record-keeping; it does all the work, I don't even look at the names of the horses before betting.

If a database has a history with containing hundreds of thousands of races I don't think one $45 horse is going to make much of an impact.

I have bet tens of thousands of races and have had good success with Access. If you are doing even better with your DSS you have my congratulations. I do think that you undersestimate what can be done with other approaches.

[hurrikane]

the best model in the world using history cannot compare with the output of a DSS using the same data in an expert system. Not opinion. Fact.
Thanks

I would tend to disagree with this. Not that a historical model is better. but you cannot issue a blanket statement that a DSS system it better and that is a 'fact'. It is your opinion.

They are two entirely different things used different ways.

[Light]

Traynor:

You sound quite skilled with computer programing,and I understand that what you have now is the best you've seen. If I understand you correctly your knock on it is that it takes up too much time.

My answer to that is what is your reward vs. your time? If I have a system making 50k a year,then I'll just consider it a job like any other and be glad to put the time into it. If it only makes 2k a year,I would feel like it's not worth my time...unless I enjoy it,which has a value too,which may be greater than money,such as relieving stress,personal enjoyment and satisfaction, etc.

I agree with Raybo 100%. If you cant find it out there,DIY. I ran into the same predicament and came up with a self made program which is the reason for my 1st place position in the Pk4 contest to date and proved itself equally profitable in last year's Pk4 contest. I dont bring this up to gloat,but to prove my point:that a DIY program, can beat just about anything out there. You certainly sound much more skilled than me with programing,so you should not have a problem developing a time saving program based on your allready successful method.

[traynor]

sjk,
My assumption is that hundreds of thousands of races is a lot more history than most handicappers use. I have no dispute with anyone using any method, nor do I have a pet preference; show me a better way and I will be the first to adopt it. That said, I am skeptical of models of hundreds of thousands of races being meaningful--that depends on how those races are categorized and modeled. The real trick is in building a meaningful model, that contains the necessary information--and ONLY the necessary information--necessary to make an optimal decision.

Whether Access or another DB, results should resemble a normally distributed bell curve; some short chalk, some longshots, most clustering around a mean. That mean should be used as the criteria, not the entire sample. Alternately, "slice" your data, extract every third result, then every seventh result, then every 13th result. If your sample is representative, the results of the slices will be essentially similar to the overall sample, and to each other.

My preference for DSS is that they factor in current reality; historical data is a factor, but not the only factor. That is, information about the entries current condition, appearance before the race, or whatever else, can be factored in. That is no more than a simple application of what Tom Ainslie called "paddock inspection" many years ago; no matter how good the entry looks on paper, if it is limping, sore, bandaged, dispirited, bored, or whatever today, in the real world, I want to base my decision on that, rather than what a thousand horses did last year or the year before.

I agree completely that mechanical methods can generate a consistent profit. That is not in contention. However, unless you know how much you are not winning by using mechanical selection, you really can't compare. If you are happy with your ROI, and you are making money, that is all that matters. If you arrived at your method using thousands of races, you are probably fairly confident that you have a big enough sample to be meaningful. I agree completely that you should be; that many races can tell you things that a few hundred races will only obscure.

Good luck, and keep up the good work.
Traynor

[SOUTHERN SLEW]

I just read your 2003 comments on the handicapping magic program, I am considering purchassing this program and would appreciate if you could share with me whether you feel this is a reliable handicapping program. If there is anyone else using this program I would appreciate any comments on it. If someone would be kind enough to post a few selections for review that would be highly recommended.

[raybo]

From Traynor: "Access is a simple database app, and I agree that it is adequate for most record keeping--as long as the input is limited to basics. That is not quite the same as DSS, the basic advantage of which is a continual, real time evaluation of results according to current reality--quite different than results based on a few anomalies ($45 winners for example) that rarely occur again. That is, the best model in the world using history cannot compare with the output of a DSS using the same data in an expert system. Not opinion. Fact."

I guess I still don't get your meaning when you speak of a DSS. You say that a DSS requires "real time evaluation of results according to current reality". What is "current reality"? Do you mean wind speed and direction, track surface moisture content, the gate being placed 10 inches off it's previous spot, whether or not a horse took a dump in the paddock, etc.? I don't know about you but the only current reality I care about is track condition, high winds and current odds, and of course the normal late changes that we all have to deal with anyway, all the rest is of an historical nature and can be gleaned from the data. That is something that would not take a terribly long time to input into a DSS program I wouldn't think.

Wouldn't a well written "AI" application do a good job of updating itself automatically and have the capability of accepting whatever "current reality" you might want to enter prior to running the reports? As far as I can tell, if you want a fully computerized handicapping program that doesn't necessarily require user input, it appears that "AI" is the future. The modeling portion of such a program would certainly be the heart of the thing, and I agree, the models must be only specific enough to guarantee similar racing conditions and physical expectations while being broad enough to get a large enough sample to assure some degree of viability.

Your statements regarding the paddock and post parade inspections are troubling, in my mind. I have personally never met a single person at the track or anywhere else that could look at the horses and tell me anything useful regarding the outcome of a race. Of course there are all the "old wives tales" but, come on----- Isn't historical data a better barometer of probable performance than a non-expert's visual inspection of unknown horses?

If you have some rules regarding inspections that haven't been disproved over and over for the past 50 years or more, I'd be interested in hearing them, although that wouldn't do me much good sitting in front of my computer watching the races on Brisbet. Whatever edge you might have in this regard would probably be negated by the distractions one must endure while being at the track, live. But, I'll listen attentively anyway.

[sjk]

Traynor,

Thanks for the encouragement. I learned a long time ago that if I start casting about for reasons not to make a bet I wind up missing out on profitable opportunities. If my computer says "bet", I bet and resist the temptation of looking at PPs, horse appearance, etc. I have my program set up so that I could make manual adjustments to the computer generated odds if I wished but in practice I do not use this.

Even if I could add a few points of ROI by looking at horses in the paddock (and I doubt that I could) I would lose out on volume. I generally bet numerous tracks and that would not be possible if I took the time to go to the paddock (or to the track for that matter; it is quicker and easier at home with the computer doing all the work).

Raybo,

I know that some others have tried the AI approach and they can report on the degree of success it has brought.

It always made sense to me to use the computer in a traditional handicapping mode. We all have a pretty good sense of the elements that come into play in handicapping a race. Rather than have the computer try to reinvent handicapping I prefer to use it to handicap as a person would but to do so in a sharper, more thorough, and more consistent (to say nothing of faster) way than I could hope to achieve on my own.

As to the issue of "current form" that you have asked about in other threads I have always used probabilities: there is x probability of running like the most recent race, y probability of running like the second last, etc.

[raybo]

From sjk: "It always made sense to me to use the computer in a traditional handicapping mode. We all have a pretty good sense of the elements that come into play in handicapping a race. Rather than have the computer try to reinvent handicapping I prefer to use it to handicap as a person would but to do so in a sharper, more thorough, and more consistent (to say nothing of faster) way than I could hope to achieve on my own."

I agree completely, and I use my program the same way you do, as a tool that takes much of the headache out of handicapping while keeping the errors to a minimum and reducing the time involved dramatically. My statement concerning "AI" was directed toward Traynor, as he was the one looking for a "hands-off" program. If one is going to use historical data to predict the future, it appears some sort of AI program eventually will do the best job of that, as it "learns" from itself via the results it produced in the past. I don't think that a viable, stand alone, hands-off AI program exists at this time that has a large enough ROI to interest me. The concept is very interesting, however, and if I was a programmer, that is what I would be working on right now.

Concerning current condition, I try to find a representative running line and then increase or decrease the grade for that race according to my evaluation of current form, rather than to apply a probability of him running that race again.

[traynor]

Raybo wrote <Your statements regarding the paddock and post parade inspections are troubling, in my mind. I have personally never met a single person at the track or anywhere else that could look at the horses and tell me anything useful regarding the outcome of a race. Of course there are all the "old wives tales" but, come on----- Isn't historical data a better barometer of probable performance than a non-expert's visual inspection of unknown horses>

Your experience is not indicative. A lot of people use inspection handicapping exclusively, and do quite well. A lot more blend it with pace handicapping and do even better. If you think that no one can "tell you anything useful regarding the outcome," you must spend way too much time at your computer and not enough at the track. Try some of the British or Australian racing forums; most handicappers in those two areas use inspection and observation as primary tools, and dismiss "number crunching" as more misleading than valuable. The myth that "everything you need to know to win" being in the DRF has sold a lot of paper. It hasn't done much else.

That also makes a point about DSS; if your pre-existing prejudice is in one direction, you will tend to ignore anything that conflicts with that prejudice. Everyone does it. It's called "heuristics."

Thanks,
Traynor

[traynor]

sjk wrote <Even if I could add a few points of ROI by looking at horses in the paddock (and I doubt that I could) I would lose out on volume. I generally bet numerous tracks and that would not be possible if I took the time to go to the paddock (or to the track for that matter; it is quicker and easier at home with the computer doing all the work).>

Yes, using a computer is easier. However, depending on the amount you bet, I disagree with "add a few point of ROI." At a track like Charlestown, Delaware Park, any track in the midwest or southwest, any track in Florida, and (best of all) Finger Lakes or Fort Erie, inspection can point out $15 to $40 winners with enough frequency to be meaningful. It is even possible to clean out the exacta pool occasionally with a fairly modest wager; that is, to take the whole pool. That is whole lot more than a few point of ROI.

It is not particularly easy, because it takes time to develop the skill, and most people cannot stand being in the position of "non-expert." All it takes is some time watching the horses, from the time they come on the track until the race is over, and comparing, taking notes, and actually studying what you see. Betting is optional.

Thanks,
Traynor

[sjk]

Traynor,

If I visited the paddock every race I could probably only play the live track. As it is I am used to playing 8-12 tracks. I don't think I could possibly make up for the lost volume.


Anyway I trust the computer handicapping far more than I would trust what I learned in the paddock (and I have spent more time inpsecting yearlings and 2YOs at sales than most handicappers).

I guess there is more than one path to the cashier's window.

[traynor]

DSS is a model of the decision making process. For some interesting reading, do a search on Google Academic or similar for "bootstrapping." It is a statistical analysis tool to model the decision making process used by "intuitive" decision makers that "reverse engineers" the decision matrix, assigns impact values to the categories and weightings the intuitive decision maker must have used to reach his or her correct decisions--and in about three iterations outperforms the intuitive decision maker. If you are not familiar with DSS, it is some genuinely spooky stuff. In arbitrary numbers, assume the intuitive decision maker selects 40% accurately; the bootstrap algorithm will "design" a model that consistently cranks out 45-50% accuracy in predictions.

I would love to run a bootstrap model on some of the old PIRCO groups races. All that "loom up off the page" stuff can be reduced to a mathematical, replicable model faster than most people can say "regression analysis."

For some interesting information (and a lot of good ideas), search the topic of decision making. It is a BIG deal in business, when decisions can involve billions, and an incorrect decision can make the difference between success and failure.

Thanks,
Traynor

[traynor]

sjk wrote <Anyway I trust the computer handicapping far more than I would trust what I learned in the paddock (and I have spent more time inpsecting yearlings and 2YOs at sales than most handicappers).>

I tend to agree, to a point. That point is reached in qualifying by inspection. I think the difference is probably that I often handicap in a group, using Delphi DSS software. (It doesn't handicap races). In the Delphi principle, a number of specialists compare choices, but each makes a prediction before he or she knows what the other group members have predicted. The decison matrix is constructed on the basis of past performance accuracy of the individual predictors, and a whole lot more. Inspection handicapping forms a very strong component of that model. It is impressive.

My personal (extremely bised and highly prejudiced) opinion is that group decision making in handicapping is a quantum leap up from individual decision making. That has nothing whatsoever to do with individual handicapping; it only involves certain key races and key situations (monster Pick Six pools being one of them). In those cases, group decision support software (GDSS) like Delphi is so far ahead of the Lone Ranger syndrome that it is not even worth comparing.
Thanks,
Traynor

[traynor]

hurrikane wrote <I would tend to disagree with this. Not that a historical model is better. but you cannot issue a blanket statement that a DSS system it better and that is a 'fact'. It is your opinion.

They are two entirely different things used different ways.>

Yes I can. I already did. It is not my opinion. Do some research on DSS, ERP, and especially the German firm SAP--that makes the software that nearly every Fortune 1000 company uses to make executive level decisions.

Thanks,
Traynor

[sjk]

Are many of them using it to bet races? How are they doing?

[sjk]

Traynor,

If your contention is that computers can out-handicap people, I would be the first to admit that my computer can probably out-do me by 30 points of ROI.

I don't see that that has much bearing on how to program the computer as to be maximally effective.

[pandy]

From my experience, both as a software-handicapping system developer and a handicapper, a good software handicapping program can help you in two ways. 1). Speeds up the process of handicapping and spotting true contenders. 2). Quickly highlights potential LIVE LONGSHOTS or overlays that most handicappers will miss. The thing is, you want to use the computer to give you a different look at the past performance information. This way you're not using the same thing (like high Beyers for instance) that everyone else is.

[raybo]

From Traynor: "Try some of the British or Australian racing forums; most handicappers in those two areas use inspection and observation as primary tools, and dismiss "number crunching" as more misleading than valuable."

Might be because they dont have many numbers to crunch. Check it out. We have much more data on horse races here than they do, I wouldn't even compare the handicapping methodologies of that type of racing and what we have over here. Yes, I would imagine that paddock inspection is a huge part of the game over there, that's about all they have available.

I personally don't go to the track unless I just want to go for the entertainment. There is much too much commotion and distraction for serious concentration.

DSS, AI, it alls seems pretty much the same to me, basically a program that learns from itself and applies that learning in a continuous upgrading of the program. Whether it is the decision making process itself or the weighting being applied to the various factors from the racing data, it's still the same logic, learn from past errors and change what caused the error, then try it again and repeat the process infinately. I do that in my own way but, of course, it is a long process. As I said before, if I were a programmer, that is exactly what I would be working on.

[traynor]

Raybo wrote <Whether it is the decision making process itself or the weighting being applied to the various factors from the racing data, it's still the same logic, learn from past errors and change what caused the error, then try it again and repeat the process infinately. I do that in my own way but, of course, it is a long process.>

Only to an extent. There are so many convoluted obfuscations (I don't get much chance to write this) in the decision making process that it is a lot more complex than it seems on the surface. For starters, how do you construct a model? What races are included, what excluded, and why? That alone is a major process. The advantage of a DSS (a good one, at least) is that it uses not just historical data, but models it in different ways. That is, process abcd might result in an ROI of 20%. Process abcd, excluding certain categories in certain circumstances, might generate substantially greater ROI, often by doing things that seem counter-intuitive.

Best example is a reduction in number of factors considered. There is a direct correlation between the confidence level of a prediction (often resulting in increased wagers, a la Beyer's "prime bets") and the number of factors considered; confidence goes up, but accuracy frequently declines. Kaywood gave a good example some years ago, using ITM% as a qualifier at minor tracks. I did the same thing with EPS in maiden claimers at Pimlico, Laurel, and Charlestown. That is, "slice" the portion of your predictions in which the top choice was also in top 3 EPS (or whatever). The combinations sometimes seem silly, but they are some of the most profitable, because few bet them.
Thanks,

[traynor]

sjk wrote <Are many of them using it to bet races? How are they doing?>

If the university I attend is any indication, handicapping is the number one effort of most MBA students, and of a substantial number of business people with access to DSS or ERP software. It is also one they find quite profitable. In one group of handicappers that I belong to, more than half the members are either current MBAs or pursuing MBAs in a university. Handicapping for business people is as much of an obsession as blackjack and poker are for computer programmers.

MBAs don't necessarily handicap better than other people, but they seem to have developed a much more professional attitude toward betting. That is, it is almost never considered as recreation, and the purpose is almost always to make a profit; just like business. They also avoid the fascination with numbers that makes people keep betting in losing situations--specifically called the "sunk cost fallacy." If you have ever dabbled with due column or other types of progressive betting, you probably know that fallacy well. Number crunching is great, saves a lot of work, but is only one part of the puzzle.
Thanks

[sjk]

Glad to hear that the university students are enjoying handicapping. I'm certainly not afraid of the competition (have a few advanced degrees myself -M.A., PhD., M.B.A.).

It is a shame if they are not enjoying the recreation aspect because it can be lots of fun.

As to the modeling process taking some time and effort I agree 100%. I have spent thousands of hours over the years in data analysis and program development.

If you take a scientific approach and invest the necessary time and effort there is every reason to expect success at this; but it sounds like you already know that.

Good luck to you.

[winbets]

A good way to try out a program for people familiar with Sartin type handicapping would be to download the Synergism VI program. The program comes with some trial races and if you are interested after using the program you must get a HDW subscription. The program itself, including all of the future upgrades, is free. The crator of the program is Bob Purdy, who was one of the original Sartinites.

[traynor]

sdk wrote <It is a shame if they are not enjoying the recreation aspect because it can be lots of fun.>

They enjoy it, but not enough to do it if they were losing. Some people do it primarily for enjoyment, and don't really care (much) if they win or lose. Las Vegas is filled with people having the time of their lives while losing money right and left. I have nothing against that at all; they enjoy it, they can afford it, and most write it off as entertainment.

I find winning enjoyable, and losing unpleasant. In order to keep at it, I pretty well have to win more than I lose; if I don't, I get disgruntled and sulk. I also stop betting. Because I like to bet (I enjoy it), I work very hard to be sure I win more than I lose. That way I can keep doing something I like, without rationalizing it as entertainment. Most of the people I know share that same philosophy--they enjoy racing, but only when it is profitable.

Thanks

[46zilzal]

Often people become SLAVISH to numbers over simple logic and will bet WHATEVER horse has the highest number, disregarding style, history at the distance or on today's surface, or "the Cramer horse" whihc always has the number but not just the winning attitude. Named in honor of Mark Cramer who first related this mania to me.

[raybo]

Traynor wrote: "There are so many convoluted obfuscations (I don't get much chance to write this) in the decision making process that it is a lot more complex than it seems on the surface."

I never said that the decision making process wasn't complex. It is extremely difficult to put the steps involved in making a decision into computer language. It's like the old tester most of us faced at some point in our education where we were tasked with trying to teach someone how to tie shoe laces, or something similar, and they weren't allowed to do anything on their own but only "exactly" what we told them. And that is relatively simple compared to all the possible "offshoots" you must consider while coding decisions concerning horse racing. This is why the user must still have input into handicapping despite using an extremely powerful piece of software. All the possible approaches to a problem make it almost impossible to code completely, like the human brain would approach the problem. It is many times more complex than the most complex coding for something like computer chess, for example, because all the possible "moves" are not always clearly defined. However, that said, coding the decision making process can be done "well enough" to be successful, albeit with some human verification of the outputs prior to wagering. I have designed my program to make decisions since day one but am constantly refining those processes, and I suspect this will continue indefinately, at least until a real "AI" program takes over that task.

[sjk]

Raybo,

I have been using my program in "black box" mode for many years (no human verification before betting). I think you will find that Access is a better platform than Excel for a thorough and complete job of handicapping and bet selection.

[raybo]

sjk wrote: "I have been using my program in "black box" mode for many years (no human verification before betting). I think you will find that Access is a better platform than Excel for a thorough and complete job of handicapping and bet selection."

sjk,

I can't argue with you concerning your point. You may, indeed, be correct. However I don't think in "Excel" or in VB or any other computer language. I think in human terms and then try to imagine how I would translate that process into computer logic, and the task is mind boggling. I can't imagine any "non-learning" platform coming anywhere close to performing all the operations that my mind goes through in analyzing even a single horse, much less a complete race, to say nothing of a complete card or a complete meet, with all the changing variables, and "constants", in racing. I doubt, if you were entirely honest with yourself, that you could even bring to conscious mind everything your brain analyzes, much less put it "all" into some kind of computer language. Now, can you do a good enough job of bringing it into conscious mind to code an application well enough to be successful? The answer, most assuredly, is "Yes". The particular platform you choose to entrust with this "intelligence" is one of personal preference and viability dependent on one's education and experience. I have never taken a computer course, except those dealing with technical procedures and troubleshooting while in the Air Force for 8 years as an avionics technician, involving the hardware side of the industry. Although, I have done quite a bit of research into the software side on my own. I have read some relational database theory and VB theory and SQL theory etc., so I am familiar, but I don't pretend to have enough knowledge to be able to use that theory in real life. I do, however, apply what knowledge I have accumulated on those subjects, and many others, in what I am doing with Excel. I chose Excel because, other than "Basic", it was the only vehicle I had available to me at the time I decided that a self-made computer application could enable me to succeed at thoroughbred handicapping. I don't doubt there are much better vehicles out there, there is no argument on that. I am currently trying to learn enough about Access to enable me to transfer some of my Excel work into a database environment in hopes of providing some more analysis power that I desire and that I can't see myself ever attaining with Excel.

The "fact" that you have been using your "blackbox" for many years is a mute point. Your application of what you have designed is, no doubt, entirely different than mine. If I were only interested in wagering on win, place or show, or pick 3 or 4, or even possibly exacta/quinella, then a "blackbox" might be viable. However, until you have tackled superfectas for 20+ years, like I have, I doubt you can make much of a case for the "blackbox" concept being successful under those conditions, if that in fact is what you are suggesting. I don't know if that was your intent and, if not, then we have no argument concerning the abilities of Access over Excel. I have no doubt that Access is a more powerful platform for storing, retrieving, and even calculating or logic arguments. However, to abandon the 20+ years I have put into my Excel application seems unfounded and careless, in my opinion.

If, on the other hand, you can teach me Access and database design in a couple weeks or months I would be delighted.

[sjk]

Raybo,

I gave up teaching a number of years ago but I have no doubt that if you dig into Access you will be able to learn to use it on your own. I think you will find that its ability to analyze dozens or hundreds of elements and then combine them is pretty much unlimited. The big advantage over Excel as I see it is that you can study a race from one point of view and store your results in a table, then move to another part of the analysis, and so on. At the end you bring everything together.

I could never handicap a race as thoroughly as my computer does. If there is anything that I as a human could do in terms of analyzing the data but that I could not program the computer to do as well or better, I am at a loss as to what it would be.

Anyway if you don't find it as useful as I have you're doing well with your Excel program.

[raybo]

Re: "The big advantage over Excel as I see it is that you can study a race from one point of view and store your results in a table, then move to another part of the analysis, and so on. At the end you bring everything together."

I understand what you are saying and I agree that creating tables and then being able to merge or combine portions of each table in database form is advantagious, however, that is also possible in Excel, albeit much more cumbersome. My Excel application does all those comparisons you speak of from different "points of view", also, and then combines the separate points of view into a suggested wagering scenario. (Currently, minus any kind of racing data of any kind just the underlying "code" and formatting, my program contains over 14 MB. When you add the database data and race card data and results data, it is quite large.) The only override I do manually concerns the running line selection made by my application. I have procedures that the program follows in selecting a likely PP but hard fast rules in this regard are only viable to a certain extent. "PP selection", as I call it, at times is highly "intuitive" in nature and expecting a machine to be as correct as I can be operating outside the "number crunching" environment of normal handicapping analysis, is highly speculative. Can it be effective enough to make a profit, probably, I have run mine "blind" over the course of an entire meet at Sam Houston and showed a healthy profit, much more than any poster I have read of here. Being able to do that over time and at different tracks "in the blind" is unsure. My only interest in handicapping is in making as much net profit as possible. I am not in it for the entertainment value or the "love of the game". It is pretty much drudgery and in the end a "real world job". I have no other income so this is serious business for me. I will probably never trust any program to make my wagering decisions for me entirely, because I will never be able to put into any program what I can do mentally and intuitively. I am constantly evolving my procedures, both "in" Excel and "out". Database design and implementation, at 55 years of age with no prior experience, seems almost a waste of my time. However, I am attempting to learn some of it, even though after "digging into" it I am more confused than before. I asked for and received, from a fellow user here, a sample database utilyzing Access. After looking at it for several days, I know no more now than I did before receiving it. I just don't know if the immense amount of time that would be required to develop a database program that would do what I want it to do would be worth getting started on, at my age and after all i have already accomplished with Excel. I have a "database" of race data and results in Excel that I run through my handicapping program to enable me to make decisions on my handicapping methods. It takes about 45 minutes to run the entire database. Your Access program is undoubtably faster than that, however, if it doesn't "learn" from itself, I doubt it is doing a better job of analysis than mine. If you still have to manually adjust your procedures involved in upgrading your program then you are still doing the same thing I have to do, just in a different environment and maybe a little faster. That said, I am still interested in the database approach, despite being embarrassingly intimidated by it. I don't hold much hope of learning how to do it, unfortunately. I feel that I am capable of much more on the software side of the problem, but starting from scratch with database design and all it would entail concerning what I envision it accomplishing, is daunting, to say the least.

[sjk]

Raybo,

I was interested to read your post. I wish you continued success.

[traynor]

Raybo wrote <If, on the other hand, you can teach me Access and database design in a couple weeks or months I would be delighted.>

Try the "Troubleshooting Access (whatever version)" and the Step-by-Step series from Microsoft. the "How to do Everything With ... " series is also great, as is the "Complete Idiot's Guide to ..."

Any of the above are worlds better than the For Dummies similar or any standard texts on the topic. If you like following instructions one keystroke at a time, try the Shelly Cashman series, used as texts in most CIS and MIS classes in the United States.

Both Excel and Access use VBA, so there are more similarities than differences. If you understand and use Excel macros, Access should not be hard for you to pick up. It is a lot easier to do it for a specific purpose than driveling through a semester long class doing mindless "projects."
Good luck!

[traynor]

Raybo wrote <Database design and implementation, at 55 years of age with no prior experience, seems almost a waste of my time.>

Databases are not hard; it is simply that most people who write about them are either incompetent, illiterate, or don't want you to learn. Do a search on Google for "database basic tutorials." There are a LOT of brief, easy to read explanations out there that will get you past the intimidation threshold. After that it is all easy.

I think www.melonfire.com (http://www.melonfire.com), then click community, and then Trog archives has some good stuff on databases and data modeling that will get you off to a flying start.

The world is made complex by people who fear sharing their knowledge. It should not be so.
Thanks