Reasons Handicapping is Tough

[thoroughbred]

I got to thinking about how tough a business handicapping is, which led me to think about important parameters that we either don't know, or know very poorly.

I'll list some here, and I ask that others join in with what they deem are important items that we can't get a good handle on.

For starters:

We don't really know the intention of the trainer, or what instructions he gave to the jockey.

We really don't know how accurate the "lengths behind" figure is. I believe it's "eyeballed" by a human. BTW, there also does not seem to be a consenus of just how long a "length" is.

The condition of the track isn't really known, in spite of the few futile attempts to let us know how deeply it was tilled, etc.

I understand that some workouts are clocked in "midstream" so to speak, i.e, the horse during the workout may have run a longer distance, and a human clocker chooses an interval of the overall distance to record the time.

Speaking of workouts, we don't know the weight of the rider during the work out.

We can only surmise the condition, or health of the horse. I understand that in Japan, the weight of the horse has to be announced, and some use it as an indicator of a horse's present condition.

As was discussed on this board by others, the race clock starts when the leading horse reaches the timing point after race run up. As was discussued by others here, that leads to errors in the time of the race. These errors can be large.

In human races, the starting points, for each racer is adjusted to take account of the racing wide around a turn. In horse racing, of course, all start from the same point, and the distance that the horse actually races, is a strong function of "racing luck." This can be a large error, because every one "width" distance from the rail adds about a length to the horses distance, e.g, a horse 5 widths out from the rail, when going around a turn, adds about 5 lengths to his total distance, or approximately an extra second. No way to know this in advance of the race.

Isn't it amazing that with all these unkowns, some can actually do well at handicapping. I'm always impressed by that.

Please add to this list of unknowns.

[cj]

-The effect of weight on a horse.

-How the horses will break...very key part of the race overlooked by most and extremely unpredictable.

-First time starters

-Horse's health

...just a few off the top of my head

[Fastracehorse]

Trainer intention is a very important part of my game. There are some very powerful hints.

Once you are good at predicting when the trainer is asking his charge to try, you have won half the battle.

Don't look at the small things is my advice.

I need to know my horse is trying and if he can run with these. How will he trip??? - difficult to know for sure but it's not to difficult to find a few good set-ups.

fffastt

[Dave Schwartz]

Handicapping is not tough. It's actually quite easy.

It is the winning part that is tough. <G>


Dave

[Pace Cap'n]

Assuming that one had the answers to the unknown variables listed so far, there is still no way to predict which horse just happens to be FEELING HIS OATS today.

[DJofSD]

Dave, you beat me to the punch (line).

DJofSD

[BillW]

Dave's comment brings up an interesting question. If we had better insight into the parameters already listed, it would probably make selecting winners easier, but how would it impact profitability? Would prices drop to impossible levels, or would the same percentage of people that are proficient in formulating wagers still have an edge? or somewhere in-between?

Bill

[Zaf]

I agree with CJ about the break. Best laid plans gone to naught all because of a bad break. D'ont you hate that.

ZAFONIC

[Kentucky Bred]

Thoroughbred, great thread and all true.


We don't know what DRUGS the horse has been given and by whom or were they administered legally or illegally. (Although I'm not sure with the new drug legislations hitting tracks around the country there is much of a difference any more.)


Kentucky Bred

[Rexdale You]

Hey Thoro. Tell us you did not forget to include the

approx 30 classes of competition MSW,Alw,Cl etc,

the various dist also must be considered,and of course

breeding as to dist & surface, I think there are a few more

But they are loading the gate and oh well next race i,ll get it

Can computer programs maybe fit it in???,,,Reminds me of a

proverb " I got where i am because of luck, and the harder i

work the luckier i get,,,,good luck to all







:confused: :confused: :confused: :confused: :confused:

[brdman12]

My pet peeve...poor jockey performance....:D

[VetScratch]

As a quasi-formal discipline, handicapping has come a long way in the past 40 years... unfortunately, the horses are still outwitting the handicappers. In the evening, after the trainer locks the tack room and the last groom leaves the shed row, the horses love to swap stories and give each other ear-wiggle signals like we exchange high-5s. :) :)

[so.cal.fan]

Handicapping may have gotten more sophisticated in the past 40 years, but what amazes me is that the favorite percentage is about the same as it was in 1903.
This varies of course with size of fields, but if you took all the 8 horse fields from 1903 and all the 8 horse fields from 2003, wouldn't the percentage of favorites be the same? I think they would be.
Thoroughbred sure brings up an interesting topic here, and I think you guys have covered it very well.
In my mind, two things never change..........you have to correctly classify the horses, and you have to correctly judge their condition right before the race. This was true in 1903 and it is true today.
Two things brought up, could be changed.........they could get a better system of correctly recording the fractional times.....and the exact lengths......install cameras at every point of call.....
They could also weigh horses before races for us, like they do in Japan......believe me on this one.......it would help.
I have seen horses that have lost approx. 200 lbs since their last start......some have been made favorites........I have NEVER seen one WIN.

[cj]

I think the favorite percentage stays about the same for a reason...the handicappers have a preference. Some will bet the chalk no matter what. Some wouldn't bet Secratariat in a Starter Allowance race. Doesn't matter how much info you have, some bet the chalk and lower price, some bet middle of the road types, others bet the bombs. That won't change no matter how "smart" we get.

[Dave Schwartz]

On the favorites' win percentage... Be aware that it is not a constant 33% all over the world. In Japn/Hong KOng it is more like 45%, in spite of huge fields.

Don't know what this is worth but just wanted to make sure that nobody thjinks 33% is some magical number.

Dave Schwartz

[so.cal.fan]

That is really interesting, Dave.
I didn't know that about Japan and Hong Kong.
It is a fact, however, that it has remained somewhat constant here in the US.

[VetScratch]

Originally posted by Dave Schwartz
Japn/Hong KOng it is more like 45%, in spite of huge fields.
Do you have a theory to account for it such as a difference in the way races are written and filled?

[Tom]

2-horse fields????:confused:

[Dave Schwartz]

More like 20-horse fields.

Although my knowledge of racing in Hong Kong is very limited by compoarison to U.S. racing, I do know that they have an "enclosed system." That is, the horses do not ship (except between the two tracks) and the races are based upon a handicap system of weights and levels.

However, in support of SOME constant, I have heard that the win percentage has been relatively stable at 45%.

Understand that this is re-gurgitated information that has come to me from other than my personal experience. If someone came forward and told me that the win perccentage was actually 65%, I would have no data to disagree beyond this.


Dave Schwartz

[VetScratch]

Well then, it sounds like 33% is another case of Muslim treachery since we do our handicapping figures in Arabic numerals! :)

[Dave Schwartz]

They only invented the numbers... Blame Bonacci.

[Seabiscuit@AR]

Dave

I have never followed HK racing in close detail but I know people who do. I have asked them how accurate the betting market is and their reply was that it was pretty similar to elsewhere (ie 30% or 33% favs). One person who follows HK all the time told me that the favs are often at $4 or $5. The fields in HK are generally 14 runners. I haven't got any precise stats but would be very surprised if the fav win percentage in HK is abnormally high.

Now I have had something to do with Japan although not for about 2 years or so. I actually did do a test on a small sample in Japan to see what the fav win percentage was. And it was higher than normal despite (as you say) very large fields. However it was not as high as 45%. It was around 37%. I remember thinking that it was an impressive figure. The thing is though that in Japan you get a lot of very short priced favourites who are long odds on. So the ROI on favs in Japan was not so great despite the great win strike rate. ROI was a loss of 17.50% on turnover. (Tote takeout in Japan is pretty high and is over 20%). One thing I remember about Japan is that I have never seen so many horses get up at 100-1 plus. Some races you had 2 or 3 favs completely dominating the market despite a large field. This meant there were numerous longshots and sometimes they get home.

I have no hard data at my fingertips for HK but I am pretty sure the fav percentage is not 45%. In Japan I think you will find the fav win % is higher than 33% but the extreme longshot percentage is also higher than elsewhere in the world. Strange but true.

[VetScratch]

Dave,
Originally posted by Dave Schwartz
They only invented the numbers... Blame Bonacci.
Are you sure that you are not confusing Blame Bonacci, an eminent lawmaker, with his brother Fib? :) :)

[Dave Schwartz]

Seabiscuit,

HK payoffs are for $5 bets.



Dave

[VetScratch]

Seabiscuit,
So the ROI on favs in Japan was not so great despite the great win strike rate. ROI was a loss of 17.50% on turnover. (Tote takeout in Japan is pretty high and is over 20%).I have seen even higher loss rates for meets at HOL and SA where takeout is less but fields are shorter than Japan. I identify favorites from final odds (e.g., 1.80) rather than win pool dollars, which produces some ties for favoritism. Using this method, I have seen negative HOL/SA ROIs greater than -20%. Maybe others who use win pool dollars to identify favorites won't agree, but I think their negative ROIs will still be at least -17.5%.

Example:
HOL 2001
Races: 841
Fav-Win%: 33.1 (as a percentage of wagers).
$2Net: 1.54 (betting all favorites in all races).
ROI: -23%
All results taken from BRIS XRD files.

[Seabiscuit@AR]

Dave

The bet unit does not matter. When I said $4 or $5 I was referring to the odds in decimal terms (ie 3-1 or 4-1). I have no doubt when I was discussing this issue with a person who follows HK all the time that he told me that favs often are at odds of 3-1 or 4-1. A second person who does HK all the time said the betting market behaviour was not radically different from elsewhere.

Japan, however, is somewhat different. But 45% favs? I don't think so. It is not my experience of Japan.

[Dave Schwartz]

Seabiscuit,

As I said, I was just re-gurgitating info that has come from our Asian clients and perhaps I have it wrong. I have never written software for them... just consulted on A.I. issues.

I could certainly have it wrong.

When I was in Japan last year, I never made it to the race track... only the betting shops. Frankly, I was in such awe at the magnitude of the place that I hardly noticed the races themselves.

Dave

[Dave Schwartz]

From this link...

http://www.hongkongjockeyclub.com/english/racing/draw.asp

... it would appear you are correct.


Dave

[formula_2002]

Originally posted by VetScratch
Example:
HOL 2001
Races: 841
Fav-Win%: 33.1 (as a percentage of wagers).
$2Net: 1.54 (betting all favorites in all races).
ROI: -23%
All results taken from BRIS XRD files.

vet , would you calculate the following for the above data.

1. sum 1/(odds+1) for all running favorites.
2. sum winning favorites (per the above that should be about 278.)

We should find that the ratio of item 2 to 1 will be must closer to the track take-out.

Thanks
Joe M

[VetScratch]

Originally posted by formula_2002
vet , would you calculate the following for the above data.

1. sum 1/(odds+1) for all running favorites.
2. sum winning favorites (per the above that should be about 278.)

We should find that the ratio of item 2 to 1 will be must closer to the track take-out.

Thanks
Joe M Sure, here are the details & results.
HOL 2001
Races: 841
Favorites: 861 (includes final-odds ties between separate wagering interests)
Wagers: 861 (betting each favored wagering interest)
Winning Payoffs: 285
Fav-Win%: 33.1 (as a percentage of wagers, or 285/861).
$2Net: 1.54
ROI: -23%

Your Requested Item-1: 352.4719
Your Requested Item-2: 285
Ratio 285/352.4719=.8086

Note: for your requested Item-1, I actually summed (1/(odds+1.05)) for all favored wagering interests.

Did you expect the ratio would be closer to .84?

[formula_2002]

VET. yes, it should get, as it did, closer to .84.

It looks like you have not eliminates the entries. When entries are eliminated you should get even closer to .84.

Thanks

Joe M

[VetScratch]

Joe,

I did eliminate the coupled-entry favorites.

Races: 841
Favorites: 861 (includes final-odds ties between separate wagering interests)
Wagers: 861 (betting each favored wagering interest)

Had I not single-counted the entries, favorites would have numbered 890.

What I think you were looking for was an answer from a method where favortism is based on win pool dollars, which would find ties only if exactly the same dollar amount was bet on different wagering interests. This was clearly NOT the method described in my original post, but for what it's worth, you got your answers in accordance with the method I posted.

My method presumes the question is, "What if I had bet all final-odds favorites?" Thus, I include ties since the player couldn't possibly break ties on the basis of pool dollars still coming in when the gate opens. Neither is this to say that the question is very practical, since final odds are hard to know before a race.

If the question is, "What if I had bet all toteboard favorites?" then even more ties come into play because of the way toteboards display floor odds for various odds-ranges.

In reality, most players probably mean to ask, "What if I had bet all toteboard favorites but skipped races with toteboard co-favorites?" When someone posts statistics based on this method, we can ask them to answer further questions.

Do you have any other self-fulfilling prophesies that you want to pose as questions?:)

[formula_2002]

vet.. I'M not looking for a method.

It's just that the 23% dollar loss betting favorites may lead some to think that favorites were not winning their acceptable number of races.

I find that the ratio of expected to actual winners is a better indication of how well horses are performing in various odds ranges.

If I think any odds range (or system) appears to be "out of order"
I'll examine each odds range, determine the standard deviation, 'z"-score and determine the probability of such “apparent "out of order".

If I have any questions after all of that, I'll use my random number generator and do some additional analysis.

Joe M

[VetScratch]

Joe,
It's just that the 23% dollar loss betting favorites may lead some to think that favorites were not winning their acceptable number of races.Yes, it always makes them feel better, as they wait in line for free soup, to know that an acceptable number of winning favorites littered their paths to destitution.:)

What they should realize from a 23% dollar loss betting favorites is that a track which produces a lot of underlaid chalk is conversely creating plenty of overlay opportunities for value players.

[formula_2002]

Originally posted by VetScratch
Joe,
Yes, it always makes them feel better, as they wait in line for free soup, to know that an acceptable number of winning favorites littered their paths to destitution.:)



Vet, In this case it's "acceptable" because its can be expected.
But if "They" dont come to undersatnd what can be expected, then they had better get a monthly pass to that soup line.

Joe M
Joe M

[VetScratch]

Joe,

If we accept that betting every race is unwise, but many recreational horseplayers do it anyway, I think the best thing for "them" to understand is that a negative ROI equal to takeout plus breakage determines whether they are handicapping better or worse than the public betting handle.

A handicapping system, like betting all favorites in our HOL example, which loses at the rate of 23% where takeout plus breakage is roughly 16.5% is performing worse than the public betting handle which is truly losing at the rate of 16.5%.

The Achilles Heel of any system that plays all races is that in some percentage of races all bets are losing propositions because the betting public has accurately handicapped the race from top to bottom. That is, if all horses are perfectly handicapped, all bets are equally attractive and will produce a loss equal to takeout and breakage when many races are considered. When the public does a "fairly accurate job" of handicapping, there may well be no winning propositions because of takeout and breakage.

This is why, unless we are having a semantics problem, that I don't understand the practical value of your statement: "If 'They' don't come to understand what can be expected, then they had better get a monthly pass to that soup line." I say this because "what can be expected" (.835) should lose at the rate of 16.5% whereas "what was" lost at the rate of 23%. In both cases, the soup line is inevitable.

I think this dilemma is what Dave was alluding to by saying: "Handicapping is not tough. It's actually quite easy. It is the winning part that is tough."

[formula_2002]

Originally posted by VetScratch
Joe,

A handicapping system, like betting all favorites in our HOL example, which loses at the rate of 23% where takeout plus breakage is roughly 16.5% is performing worse than the public betting handle which is truly losing at the rate of 16.5%.


Vet, its simple point but it’s the only one I really want to make here.
I would not compare the 23% loss (a flat bet play) to the take-out. I want to compare the results of a Dutch bet to the take-out (1/(odds+1)).
That sum, in this case was .81% which, for purposes of comparison would be a 19% loss vs. the 23% loss.

As you know, take-out and breakage can be calculated for any track by summing 1/(odds+1) for all horses. The reciprocal of that sum -1 =take out.
It’s just comparing apples to apples..

[VetScratch]

Joe,
As you know, take-out and breakage can be calculated for any track by summing 1/(odds+1) for all horses. The reciprocal of that sum -1 =take out.Believe me, I certainly sensed that you are knowledgeable when you asked for the calculations, which is why I used the sum of 1/(odds+1.05) to compensate for breakage.

There is nothing to refute about what you have posted. My implication was that the "general" audience for horseracing posts does not save the necessary data to analyze the pari-mutuels. Many casual horseplayers "feel" their losses but don't even keep the records required to analyze their personal ROI for various odd-ranges.

[formula_2002]

Originally posted by VetScratch

My implication was that the "general" audience for horseracing posts does not save the necessary data to analyze the pari-mutuels. Many casual horseplayers "feel" their losses but don't even keep the records required to analyze their personal ROI for various odd-ranges.

Yep, you’re right about that.
And it goes even further... I think, given the volumes of data and records many knowledgeable players have, there is little or no regard for proving the significance of their play.

A lot of winning that happens by chance is taken for a long term profitable system. Worst, long term profits are not even a consideration.
It' not the handicapping or the winning that's difficult, it's the complete understating of the SIGIFICANCE of everything within the system!

So much data is required to prove a system, I'm not certain that it has or can be done (the proof that is).

I haven't met anyone that agrees with that statement, but I also have not met anyone that can demonstrate it, first hand, to be wrong.

[VetScratch]

Joe,
And it goes even further... I think, given the volumes of data and records many knowledgeable players have, there is little or no regard for proving the significance of their play.Quite true... I have an otherwise brilliant friend in Minnesota whose idea of monitoring his play is to check his barns every morning to make sure he still has his gull-wing Mercedes and other toys! :)

[P. D. Mahalik]

The more I read the more I know how little we all know about that data that is out there, but no one is willing to mine it. That is why I HAD to develop MMTRX (BRISnet.com Version). The DRF doesn't show turn time in feet per second and what Brohamer/Sartin call the most important element of handicapping -the second call in feet per second. It is all there, but no one wants to crunch all of those numbers. Now at least I have that at my finger tips for a dollar BRIS file. This is the most exciting thing I have done in years.

[thoroughbred]

Originally posted by P. D. Mahalik
The more I read the more I know how little we all know about that data that is out there, but no one is willing to mine it. That is why I HAD to develop MMTRX (BRISnet.com Version). The DRF doesn't show turn time in feet per second and what Brohamer/Sartin call the most important element of handicapping -the second call in feet per second. It is all there, but no one wants to crunch all of those numbers. Now at least I have that at my finger tips for a dollar BRIS file. This is the most exciting thing I have done in years.

Do you have any insight as to why turn time is the most important element of handicapping?

With all the factors out there, early speed, ability to rate, etc. what is the reasoning behind turn time being so important?

[cj]

PH doesn't say turn time is most important, he says the second call is according to the Sartin guys. Of course, it is readily available in Formulator with a few setting changes, just not in FPS. No matter how you measure it, time is time is time...

[P. D. Mahalik]

Turn-time, second call really, is so important because this is the segment of the race where everything comes together. Decisions are made based on how well the horse is doing in this segment. By knowing the second call feet per second from the last race, we get some clue as to what might happen this race.

[cj]

No argument here on the importance of the second call. I'm just saying it doesn't matter how you measure it. It can be feet per second, miles per hour, actual clock time, seconds per hundred yards, or converted to a pace figure...its all the same measurement. Just a matter of what form of the measurement you are comfortable using.

[rmania]

Originally posted by P. D. Mahalik
Turn-time, second call really, is so important because this is the segment of the race where everything comes together. Decisions are made based on how well the horse is doing in this segment. By knowing the second call feet per second from the last race, we get some clue as to what might happen this race.

Hmmm...

Maybe this is why some methods are more successful when zeroing in on this part of the race. :rolleyes:

[Fastracehorse]

There is one reason why turn times are important - what is it??

It's not rocket science either - it is very fundamental - very elementary.

:rolleyes: :confused: :rolleyes: :rolleyes: :rolleyes: :rolleyes:

fffastt

[kenwoodallpromos]

Horses with energy still in the tank usually do well on the last turn; horses trained well will change leads correctly; a lot of the jockeys push the button about then; and if there is an effecting track bias many times the bias on the rail can start affecting the leaders in the turn.

[kenwoodallpromos]

I'm Always looking for new mthods! P.h.- can I email concerning my track speed (variant) prediction method using track's daily 4f works?

[Tom]

Originally posted by Fastracehorse@DRF
There is one reason why turn times are important - what is it??

It's not rocket science either - it is very fundamental - very elementary.

:rolleyes: :confused: :rolleyes: :rolleyes: :rolleyes: :rolleyes:

fffastt

Becasue if you don't run the turn time, you can't run the final fraction? :p

[keilan]

Much more is involved than simply isolating the horse with the best velocity in the middle fraction. Horses that are unable to run two consecutive fractions are generally poor wagers. This, like most calculations are not stand alone variables.

However horses that run this fraction on the turn in an efficient manner demonstrate the characteristic's of a fit horse. Further to this turn times are more reliable in sprints than routes.

[Fastracehorse]

Thanx for your input.

Can you please expand on the point, "turn times are less important in routes."

And, what part of the United States do you live in??

Thanx,

fffastt

[cj]

He said they are more reliable in sprints, not less important in routes. There is a much wider range of pace scenarios in routes. The horses have already traveled a half mile as opposed to a quarter.

I also think turn times don't mean much without a variant applied to them, or if comparing unlike distances.

[keilan]

CJ --you are quite right about the different pace scenarios and unlike distances seen in routes. This coupled with the track configuration, distance run between 1st and 2nd call is often on the back stretch etc.

Routes IMO have a different set of criteria/variables than those used for sprint handicapping. Simply stated turn time is less significant in routes as other factors become more important. As was previously stated "turn times are more reliable in sprints than routes".

Those that have calculated their own measurements understand the importance of accurate par times and variants.

Fastt it's really very fundamental - very elementary.

Enjoy the rest of the day!

[Fastracehorse]

You said this:'He said they are more reliable in sprints, not less important in routes. '

=================

Then Keilan said:'Simply stated turn time is less significant in routes as other factors become more important. '

==========================

I'm confused - OK I'm not - I was just funning with the cynical question to Keilan.

========================

Now, here's what I think: Turn times aren't important unless one vital factor is determined - lost ground. You can't lose ground anywhere else!!

Speed actually carries better in routes than in sprints - McCarron believed this was true in turf routes as well - the longer the better for speed. In routes, there is a better chance of thievery - bottleing up the competition and hence, making turn times at least as important in routes than in sprints.

My adjusted figure counts turn times equal to both sprint and route events.

fffastt:cool: :o :) :D :eek:

[keilan]

fffast -- huh!

<Turn times aren't important unless one vital factor is determined - lost ground. You can't lose ground anywhere else!!

What happens in the event that the turn time calculation doesn't occur on the turn? Is being 3 wide down the backstretch the same as being 3 wide on the turn? Do you think so? What if the 1st fraction went in a pedestrian pace?

Losing ground on the turn isn't necessarily the same portion of the track where Turn Time is calculated. No -- Sounds like you are using them interchangeably.

Please share your thoughts -- I'm interested.

[Fastracehorse]

Keilan: Is being 3 wide down the backstretch the same as being 3 wide on the turn?


fffastt: Turn times aren't important unless one vital factor is determined - lost ground. You can't lose ground anywhere else!!

===========================

I think you answered your own question already.

=====================================

Keilan:Losing ground on the turn isn't necessarily the same portion of the track where Turn Time is calculated. No -- Sounds like you are using them interchangeably

fffastt: In my figs it's the only time when TT is calculated ( lost ground ) - I don't worry about TT per se.

fffastt

[DJofSD]

OK, I think I get it. TT is only important when it's important.

DJofSD

[freeneasy]

Originally posted by so.cal.fan
Handicapping may have gotten more sophisticated in the past 40 years, but what amazes me is that the favorite percentage is about the same as it was in 1903.
This varies of course with size of fields, but if you took all the 8 horse fields from 1903 and all the 8 horse fields from 2003, wouldn't the percentage of favorites be the same? I think they would be.
Thoroughbred sure brings up an interesting topic here, and I think you guys have covered it very well.
In my mind, two things never change..........you have to correctly classify the horses, and you have to correctly judge their condition right before the race. This was true in 1903 and it is true today.
Two things brought up, could be changed.........they could get a better system of correctly recording the fractional times.....and the exact lengths......install cameras at every point of call.....
They could also weigh horses before races for us, like they do in Japan......believe me on this one.......it would help.
I have seen horses that have lost approx. 200 lbs since their last start......some have been made favorites........I have NEVER seen one WIN.

yes so. cal. i couldnt agree with you more espeacially on something that can be as important and varifiably so as the weight issue, loss/gain, you touched on. ive been aware of this type factor for some time and believe that it is a factor that many, many handicappers would base their final decision upon. to bet or not to bet.
possibly when a horse is enterd then he must be weighed, and that in ample enough time, oh say 3 to 4 to 5 days before his race day, in order to allow publications to have this information printed by the time they go to press. would be nice.

and just for my own reasoning so. cal, what would you evaluate as a normal and/or reasonable weight loss?
some horses lose weight after racing and thats normal, but when a horse does not regain that lost weight, then where do you think an acceptable / unaceptable peramiter of weight loss would lay. this is a very interesting subject and one that has hardly ever been touched on.

[kenwoodallpromos]

Last time I looked, online Hong Kong Racing gave total weight.

[formula_2002]

three reasons that make handicapping tough.
1. not keep score properly.
2. finding correlations where there are none
3. not having a sufficient amount of data.

Joe M

[Fastracehorse]

Whoops you forgot #4.

4. Haveing too much non-contextualized data.

fffastt