i need some help in making my own line

[PSTfredk]

I read cramer book on value handicapping and it is very confusing
he say that you line always has to add up to 100.
in the first race at oaklawn i added up the m/l odds and
they equal 88. how do i adjust the odds to total 100.

[kenwoodallpromos]

Sorry- I was confused too! the boook "a day at the races" N.Cal's program odds guy Richard Somers tells, says add up to 100 and add takeout; more good info by him and other track people. I decided it was easier to look for horses that were overlays for race condition so I do not make lines- just try to see when a horse is longer odds than they should be and why bettors favor some factors over others. I find some good overlays at say 20-1 that should be 4-1!

[Buddha]

How did you get them to add to 88? If you add the percentage according to the odds, it is 127+%.

http://www.todaysracingdigest.com/freearea/Handitips/betline.htm

That is the website that I have used to help me make betting lines when I have tried to make some. If you have any questions, I think I understand the process good enough that I can help. You can either PM me, email me, or post here.

[JackS]

Making a line to include every horse in the race is a time consuming endevour. An easier way is to pick 4-5 (or more if judged to be "wide open")logical contenders. Assign odds based on your own ability to pick horses and to place them in order of your preference. (1,2,3,4,5etc) If your top choice has a win % of 28, fair odds would be about 5/2 or break even. Your 2nd 3rd and 4th choice should have corrospondingly higher odds. This level is measured on actual races handycapped and a minium of 10-20 races should be enough to get you started. Monitor your abilities frequently and change your line accordingly. Don't forget the track take and look for overlays within your line.

[Exactaman]

Originally posted by Buddha
How did you get them to add to 88? If you add the percentage according to the odds, it is 127+%.



a m/l could add up to 88, although it would be a very bad ml. sure you didn't slip up PST?

PST, to get to a hundred from there, you have 12 points left. that's about 1 point for each 8% in the line. just add points to the other horses on this ratio, as best you can.

[PSTfredk]

i assume i used the wrong thing to start with. i need to change the odds to percent chance of winning, from the answers i got back i assume the correct percent would be 127, if thats the case how do i reduce that number to 100 to start with.

[Brian Flewwelling]

Originally posted by PSTfredk
...how do i reduce that number to 100 to start with.

divide each number by 1.27.

[JackS]

If the ML in a 10 horse race equals 127% ,increase ML odds by 2.7% on each horse. An easier way would be to eyeball it and raise the odds by a factor of 1.( i.e. 9/5 to 2/1, 2/1 to 5/2, 5/2 to 3/1 etc.) Exactly a 100% line is not necessary. My guess any line between 97 to 100% would be fine. The only disadvantage of going to far below a 100% line is that you may price yourself out of some nice plays. On the other hand, with patients, overlays could really "pop"on occasion.

[Buddha]

Originally posted by PSTfredk
i assume i used the wrong thing to start with. i need to change the odds to percent chance of winning, from the answers i got back i assume the correct percent would be 127, if thats the case how do i reduce that number to 100 to start with.

Well there are two different ideas behind lines. The morning line oddsmaker has to account for the track takeout, which actually wouldn't be as high as the extra 27% added to the total amount, but they use that for cushion. If you are making a betting line, I have heard that you will only want to make the line to 100% and that is it, you don't need the "cushion", so therefore, you would have to change the odds of the horses to make 100% based on your belief of their chances. It doesn't have to be exactly 100%, just close.

Also, if you ever read Steve Fierros Four Quarters To Horse Investing, he makes his betting line on an 80/20 basis. He will give his 2,3, or 4 contenders 80% of the line, and then leaves the other 20% percent for his horses he has think has no shot, and doesnt even give them a line, just his contenders.

Email me if you want more info on the Fierro way of doing it. Will be glad to try to help.

[Larry Hamilton]

That 80/20 rule has value because Steve can consistently put the winner in the top 4. If you can do better, you may be able to use a 90/10 or worse maybe you need to use a 70/30. THe point is: measure you own ability and use your own rule. In fact, if you know how well you do with this TYPE of race you can use a different set of values to match your own ability

[Valuist]

I think there's some severe problems with only making odds on the top 3-4 horses. First off, I think its a huge mistake to lump a 12-1 shot into the same boat as a 75-1 shot. The 12-1 shot may have an excellent shot at hitting the board while a true 75-1 shot would be very, very happy to get 4th. If you play tris, supers and exactas, you should make odds for all the horses. The best value in exactas and tris is those double digit odd horses in the underneath slots.

I think the best way to handle the longshots is price them first.

[Exactaman]

Originally posted by PSTfredk
i assume i used the wrong thing to start with. i need to change the odds to percent chance of winning, from the answers i got back i assume the correct percent would be 127, if thats the case how do i reduce that number to 100 to start with.

Ahhh. Just curious PST, why do you want to do this to the ml? I don't see the usefulness of it.

Originally posted by Buddha
I have heard that you will only want to make the line to 100% and that is it, you don't need the "cushion"


The idea here is that you want a pure percentage chance of your horse winning. you can then compare his to the prices on offer to make your decisions. there's no reason to adjust your line for the takeout.

[JackS]

Almost everyone trying to make their own line is going to run into problems starting with horses who have lesser chances. These problems are going to start for most of us probably around 10-15 to one. Odds lines really become subjective and more of a point of view rather than anything that would be truly useful. My suggestion for most who are intent on doing this, is to use the official ML and make percentage adjustments to this line rather than rewrite you own. There may be a few people who do understand their abilities and are successsful but, for most of us, it's probably time unwisely spent.

[Tom]

Originally posted by Larry Hamilton
That 80/20 rule has value because Steve can consistently put the winner in the top 4. If you can do better, you may be able to use a 90/10 or worse maybe you need to use a 70/30. THe point is: measure you own ability and use your own rule. In fact, if you know how well you do with this TYPE of race you can use a different set of values to match your own ability

Good point, Larry.
I use HTR and I keep track of my contender factors in the short term and make odds line adjustments accordingly. Overall, my basic method gets 80% winner in the top 4, but at some tracks and some types of races, it may do 85%, or it may only around 70%. And I can compare this to over two years worht of history if I want to.
When my factors are clicking, I am atunded to it and base my play on what is going on right now, say last 20-30 races of the type.
This also telss me what races to just stay out of until my methods are picking them.
Those templates from 4 Quarters really helped me with odds line making.

[Aussieplayer]

This is part of what I wrote someone in a PM. Maybe it'll help?

"Bear in mind, that the actual table idea is what I wanted to portray. The actual values of the increments should probably be tailored to your own figures.

I think you can make a good line with a simple 1.0, 0.9, 0.8, 0.7 etc. etc. - but what Don Scott's values do (if you think about it) is weight the line more heavily to the first few contenders, which, if you think about it - is how it should be."

Cheers
AP

[ljb]

Ok folks,
I would like to throw another thought/question into this thread.
After analysis of a race the horses are rated 1,2,3,4 and so on.
Say my top choice wins 30 percent. This translates into about 5 to 2 as break even odds on my top choice. Indicating to me I must get a minimum of 5-2 to even break even.
However upon further review I find, my top choice has won 30 percent at odds of 2-1 or less.
What now?

[JackS]

IMO you will have to avoid a win bet on your top choice at anything below 3-1. However , you could look at exacta overlays using this horse. If you have win % figures for your 2nd, 3rd ,4th choices, value may exist with one of these horses. The value of your line is exactly it's performance and the warning it's giving you. Money saved is as good as money earned.

[Aussieplayer]

ljb & JackS:

Finding that your top choice wins (say) 25% and then deciding that as long as you take more than 3/1 is totally fraught with danger, and a way to the poorhouse (in my experience). Despite what you read in books.

What is really going on is that your top choice, whilst it may AVERAGE out at (say) 25%, it will be a variety of percentages - DEPENDANT on the actual odds.

ie. Your top choice that goes off as public fav may well win at a 35% clip, but may have a negative ROI.

Your top choice at 10/1+ odds (you think you're getting a great overlay) - MAY have a positive ROI, but may just as likely have a NEGATIVE ROI.

You might find an odds range that is profitable - YOU MAY FIND THAT NONE ARE PROFITABLE!!!!!

It has more to do with the methods that you are using to come UP with the line (or set of rankings) in the first place - rather than the line itself.

(IMO) nearly any line is a dog of a line in reality (without incorporating public odds). But if your METHODS use some contrarian thoughts that ACTUALLY WORK, you THEN should find some odds that allow a profit. Rick is a prime example that we all know well. He uses a 3 factor rating method that both apparantly fulfill the vital requirements:
(a) point to un-public horses
(b) it is a valid method of winner finding

THEN - he finds that as long as he avoids the m/l fav. and top rated speed horse, he can make a $.

What I'm trying to say is that it's just as easy (in my findings) to invent some rating system that puts longshots as contenders BUT DOESN'T actually have enough of them winning to profit - as it is to come up with a rating system that uses common factors, and hence leads to chalk.

I'd like some discussion on that issue. Rick?

Anyway - back to what I started saying - don't just assume that you need greater odds than your average win percent. It doesn't work like that (necessarily).

One last thing I'd like to share is this: I have found it far better to analyse ranks by degrees of overlay, than by odds levels. Try it & see.

Cheers
AP

[JackS]

Aussie- Not quite sure I understand all you've said, but trust it to be knowledgable. The main problem as I see t with the morning line is that the lines maker is making this line in anticipation of the public money that will come in on this race. The better he is, the more accurate his prediction. These guys are very knowledgable and most quick to admit that the ML they make does not necessarily reflect the way they would choose to bet their own money. For someone setting out to discover their own abilities, I'd suggest ML adjustments in the line and also a re-ranking if necessary. Small wagers on percieved overlays and a written record of the results over a meaningful period of time. A method such as this should result in small profits or losses and at the same time allow for a little handicapping insight to develop. Like a very cold swimming pool. If you want to go swimming, you'll have to do more thn test the water with your foot. Don't jump in either, just slide in real easy.

[JustRalph]

This why I don't make an odds line. I dabbled with it, but found that it is a hell of a lot of work, and maybe not that useful. I agree with finding 3-4 contenders and then I look at other things. My Software sets me an oddsline from the database portion of the program.......after my 3-4 contenders I start looking at the oddsline from the database program for exotics etc. of course if my database oddsline has a horse that is as low as my contenders I use it to go back and make sure I didnt' miss anything. It works the other way too. If the program decides that one of my 3-4 contenders is a throw out or lets say 10-1 or greater, I look hard at that horse again. And re-evaluate my contenders. Usually I stay with what I think is right, or I add some filters etc to see the differences.

[alysheba88]

I make odds on 3-4 horses a race, never assigning higher than 6-1 and get plenty of plays at 10-1 or more, especially with exotics. Not unusual to have a 20-1 plus horse in there.

[PSTfredk]

if you never assign a horse higher than 6/1 how do you get

winners that pay 10/1 or 20/1

[GameTheory]

Because you bet them if they are higher than on your line.

Assign a horse 6/1, that means bet it at higher than 6/1...

[PSTfredk]

OK

[dav4463]

Keep a record of your picks and their winning odds. You can break it up into race types as well...routes/sprints/maiden/turf...

An example:

Your top pick shows a profit at odds of 8-5 through 9-2 and at 14-1 or higher, but shows a loss at 5-1 through 13-1, so avoid those odds.

Second pick shows profit at 4-1 through 8-1 and 14-1 or higher, but shows a loss at 9-1 through 13-1

Third pick shows a profit at 8-1 through 14-1, but shows a loss at higher than 14-1.

Fourth pick shows a profit at 14-1 or higher only.

Wait for the opportunities to bet your picks at their profitable odds only.

Review records often and change odds requirements accordingly.

[Macdiarmadillo]

JackS wrote:
The main problem as I see it with the morning line is that the lines maker is making this line in anticipation of the public money that will come in on this race. The better he is, the more accurate his prediction.

Public handicappers tend not to make favorites much under even money or longshots much over 30-1 even if it painfully obvious this should be so. You'll get all kinds of reasons, but it's basically about saving one's behind. Make some horse 99-1 and you'll hear about it from the owner. A 1-9 loses by a nose and they'll be chasing you down the street.

I agree that the work is not worth the great amount of time needed to generate odds lines. Better to study a lot of races and learn the factors that determine what odds horses can be.

[Derek2U]

I got to read that Fierro book -- yeah thats my next read -- but
I think that doing an odds line is just busy work for many players.
Sometimes that 3:5 is like 99:1 sure to win & then what ...
Pass / Bet ?