Do Horses Win Only 5% When Returning After a Scratch?

[Krudler]

I've been reading the "Has Anybody Here Heard Of…." thread and it mentioned a 2008-ish or so interview with Jim Clark, Director of (the now defunct) In-Form Handicapping System. In that interview, Clark makes the following claim:

"The percentage of horses winning off a scratch…is less than 5%, so you can pretty well toss them."

I've never heard of this dismal-next-race-after-a-scratch angle. i'm wondering if anyone else has and, if so, are the percentages that low nowadays?

If you're curious, I've attached the entire article to this message.

[Greyfox]

That hasn't been my experience.

In So Cal horses are sometimes entered two days in a row.
Depending on the card, the trainer can scratch from day 1 and stay in day 2.
I haven't kept stats, but 5 % sounds low.

[lamboguy]

i would say that number is correct from many years ago, but probably not today

[cj]

I've been reading the "Has Anybody Here Heard Of…." thread and it mentioned a 2008-ish or so interview with Jim Clark, Director of (the now defunct) In-Form Handicapping System. In that interview, Clark makes the following claim:

"The percentage of horses winning off a scratch…is less than 5%, so you can pretty well toss them."

I've never heard of this dismal-next-race-after-a-scratch angle. i'm wondering if anyone else has and, if so, are the percentages that low nowadays?

If you're curious, I've attached the entire article to this message.

Maybe he meant vet scratch?

[horses4courses]

Maybe he meant vet scratch?

Very good point.

There's obviously a huge difference between being forced to scratch by a vet due to some physical issue,
and connections picking and choosing engagements for their horse.

I would imagine that the winning percentage for vet scratches is far lower than other types. If you broke down scratches due to alternative engagements among a select group of savvy trainers, you might even get a decent winning percentage with a reasonable ROI.

[therussmeister]

As a handicapper that pays attention to vet scratches, there are two instances I can tell you where you definitely can not safely toss them.

The first is a "quick return" scratch. This is when a horse ran six or less days before being scratched. I've seen horses run one day, and then were a vet scratch the very next day at a different track. I doubt the trainer really wanted to run two days in a row, but I don't see why it had to be a vet scratch instead of a trainer scratch.

The second is when a horse has a solid five furlong workout since the scratch, or sometimes two fast four furlong workouts.

Lastly, I strongly suspect that a lot of horses that were scratched at the gate are good to go in their next start, as many gate injuries turn out to be minor. But I don't know where I could get information to confirm this hypothesis.

[Krudler]

Interesting thoughts. I too had thought Clark might be referencing vet scratches as the source of the low win & %age. Also, I have a hard time believing that a trainer scratch that moved a horse from say a Saturday to a Sunday card - likely a sign of positive trainer intent and a well meant horse - would result in such a low hit rate.

I've just doubled back to the article Informed (and Enriched) with In-Form Handicapping to reread Jim's thoughts. Here they are:

“We’ve been following this for seven or eight years and it’s a huge handicapping angle,” insists Clark, “A scratch board horse is one who has been scratched from a race previous to today’s race. It may have been a trainer scratch, steward scratch, vet scratch or off the turf. The percentage of horses winning off a scratch board is less than 5%, so you can pretty well toss them.”

Clearly, in his mind at the time of the article's writing, a scratch is a scratch is a scratch.

Seeing as Jim has just signed on as a PA member (he first post was in the "Has Anybody Here Heard Of…." thread), I plan on asking him to elaborate on this current thoughts on this angle.

[DeltaLover]

I've been reading the "Has Anybody Here Heard Of…." thread and it mentioned a 2008-ish or so interview with Jim Clark, Director of (the now defunct) In-Form Handicapping System. In that interview, Clark makes the following claim:

"The percentage of horses winning off a scratch…is less than 5%, so you can pretty well toss them."

I've never heard of this dismal-next-race-after-a-scratch angle. i'm wondering if anyone else has and, if so, are the percentages that low nowadays?

If you're curious, I've attached the entire article to this message.


The claim that horses of the scratch are only winning 5% is wrong. Obsiously the linked article is just an ad for an approach that does not work (judging from the fact that the promoted site is no longer functional). In general, I will avoid any blanket opinion like this, especially when data are not provided. More than this, note that such a low winning frequency is very difficult to attain for any kind of a handicapping factor especially when crowd's odds are not part of it.

[Saratoga_Mike]

It isn't hard to get a vet scratch - I wouldn't use that as an angle.

[Robert Goren]

At NYRA where I bet, there seems to be several reasons for Vets Scratch. If I am watching when there is gate scratch I note why. Run off, injured in gate, etc. If I see a vets scratch and I don't why, I consider it a big negative unless I have seen the trainer win with one. This is one thing that the public either doesn't know about or ignores and therefore offers potential value.

[thaskalos]

Am I wrong...or are we being advised to eliminate a horse from today's race, just because it was scratched out of a race which was taken off the turf in its prior start?

I see all kinds of scratches when races are taken off the turf; am I supposed to believe that these horses are automatic throw-outs when they show up for their next start?

The attached article here is an old sales pitch reminiscent of the system peddlers of old. The guy who wrote it wouldn't even give us his full name -- choosing to initial his article instead -- and who could blame him. He misspells Ragozin's name in a passing reference to him...takes $40 with him to the track so he can try out a new, promising method...and then is seen clutching his winnings as he is heading for the elevator -- even though he had previously told us that he had "pocketed his winnings".

I wouldn't sign my name to such an article either...

[BIG HIT]

he did bring some thing not talked about here before.I tried it on one day you guy's defintely right aft 35 plays 6 different trks had 6 winners 4 2and but i think you can make it work for instance and i have not check this yet was chg trk drop class chg jky lower chg dist and surface.He gave the in pression he was come back with more maybe he will it was interesting read

[Pensacola Pete]

It depends on who scratched the horse and how long it's been away.

Trainer scratches are often the trainer looking for a better spot.

[raybo]

Assuming that 5% is correct, there will still be portions of that 5% that will be successful next time out, just like any other blanket statement about an "overall" percentage. In every "general" percentage I can think of, there are exceptions, and to just throw them all out is probably not advisable, and could actually cost you money by doing so.

[Robert Goren]

I am surprised none of the database guys have come in with a number.

[chadk66]

no way that's the percentage now days. hell more than 5% scratch every race :D

[Triple Nickle]

I don't know if HDW tracks this info. If not,then I would guess it's not in j-Cap. HTR, HSH,All in One or any other program using HDW info. Although I could be wrong.

Best
Ken

[antigeekess]

That hasn't been my experience.

In So Cal horses are sometimes entered two days in a row.
Depending on the card, the trainer can scratch from day 1 and stay in day 2.
I haven't kept stats, but 5 % sounds low.

That was my first thought when I read that too, Greyfox -- the "enter both races, scratch one later" scenario. And 5% certainly sounds low in these instances, considering that trainer certainly is choosing the better of the two spots for the horse.

I finally figured out that I needed to keep an eye on these double-entered likely scratches when playing exotics one weekend when I ended up getting paid 3 times on a small Pick 4, due to multiple scratches and being assigned the winning favorite in each of their places. Such a deal. :)

[raybo]

I am surprised none of the database guys have come in with a number.

They would have to have race notes wouldn't they, to include what kind of scratch it was, over and above just a vet scratch notation, because scratches are defined so generally in the data files, the results of such a query would not mean much.

[jennywilsonhrs]

I have to agree with a lot of the responses already posted here. First, the 5% figure may be accurate if you consider all the different types of scratches, but personally, I would not discount a horse just because it scratched the race before. Second, I absolutely agree with raybo that there will be exceptions to the 5%.

[stuball]

I have all the scratches and a scratch is not a scratch.....as there are 9 different designations for scratches in my database....would have to break down by
type - track - trainer -etc etc etc...very labor intensive...not sure I would want to tackle this project...now maybe if my boss told me to do it....but she is getting ready for Black Friday....LOL :lol:

Stuball

[Ocala Mike]

Assuming that 5% is correct




What am I not understanding about statistics and/or the English language?

If it's ASSUMED that 5% is correct, then there will be EXACTLY one "exception" every 20 cases.

I guess raybo and jennywilsonhrs's job is to find it.

[raybo]

What am I not understanding about statistics and/or the English language?

If it's ASSUMED that 5% is correct, then there will be EXACTLY one "exception" every 20 cases.

I guess raybo and jennywilsonhrs's job is to find it.

You mis-read the meaning of the post. The percentage could be anything, and there will still be exceptions to such a general rule. My statement was that whether or not the actual percentage that win next time out is 5%, there will still be a percentage of that percentage that should not be thrown out blindly.

[Ocala Mike]

OK; there are exceptions to every rule, which, I guess, is why I don't follow systems of play that have hard and fast rules. That, of course, is a whole other topic.

[stuball]

Scratch Stats for 2011-2012-2013

TYPE SCRATCHED WINS WIN% AVWINPRICE
ALL 113,698 13879 12 12.53
VET 37594 4453 11.8 12.67
STEW 21932 2594 11.8 12.89
TRNR 22987 3093 13.9 11.40
ALSO ELIGIBLE 16540 1790 10.8 14.69
OFFTURF 6490 775 11.9 12.07
MAIN TRK ONLY 4271 684 16 10.67
DOUBLE ENTERED 3841 483 12.5 11.60
GATE 41 6 14.6 25.83


This should finish the 5% myth very few gate scratches which I presume
are lumped as a stewards scratch...why do I do this --spent the better part
of 2 hours figuring this out....oh well, wish I knoew how to post this right

Stuball

[davew]

Does anyone know what they consider a scratch to get the 5%?

Do also eligibles that did not start count? Run so track muddy and turf to crap scratch count? Part of an entry scratch count? Horse warming up and runs away in post parade scratch count?

[Robert Goren]

If I read stuball's numbers right, all groups win a lot more than 5%, but none except gate scratches showed a profit. The gate sample is pretty small and maybe one large winner skewed the average winner's price. It looks like a scratch is a pretty strong negative, but not an automatic toss. I will start watching for trainers who win after a scratch. Thanks stuball for posting.

[Greyfox]

Scratch Stats for 2011-2012-2013

TYPE SCRATCHED WINS WIN% AVWINPRICE
ALL 113,698 13879 12 12.53
VET 37594 4453 11.8 12.67
STEW 21932 2594 11.8 12.89
TRNR 22987 3093 13.9 11.40
ALSO ELIGIBLE 16540 1790 10.8 14.69
OFFTURF 6490 775 11.9 12.07
MAIN TRK ONLY 4271 684 16 10.67
DOUBLE ENTERED 3841 483 12.5 11.60
GATE 41 6 14.6 25.83


This should finish the 5% myth very few gate scratches which I presume
are lumped as a stewards scratch...why do I do this --spent the better part
of 2 hours figuring this out....oh well, wish I knoew how to post this right

Stuball

Thank you. :ThmbUp:

[raybo]

If I read stuball's numbers right, all groups win a lot more than 5%, but none except gate scratches showed a profit. The gate sample is pretty small and maybe one large winner skewed the average winner's price. It looks like a scratch is a pretty strong negative, but not an automatic toss. I will start watching for trainers who win after a scratch. Thanks stuball for posting.

That pretty much falls into what I think, regarding scratches, what is the reason for the scratch if it's a vet scratch, and what is the trainer's intent if he scratches the horse? Circumstances, and intent, are very important factors.

[VeryOldMan]

Scratch Stats for 2011-2012-2013

TYPE SCRATCHED WINS WIN% AVWINPRICE
ALL 113,698 13879 12 12.53
VET 37594 4453 11.8 12.67
STEW 21932 2594 11.8 12.89
TRNR 22987 3093 13.9 11.40
ALSO ELIGIBLE 16540 1790 10.8 14.69
OFFTURF 6490 775 11.9 12.07
MAIN TRK ONLY 4271 684 16 10.67
DOUBLE ENTERED 3841 483 12.5 11.60
GATE 41 6 14.6 25.83


This should finish the 5% myth very few gate scratches which I presume
are lumped as a stewards scratch...why do I do this --spent the better part
of 2 hours figuring this out....oh well, wish I knoew how to post this right

Stuball

Wow - very nice contribution! Should end the "5%" discussion once and for all. Whether it reveals anything statistically significant about the different types of scratches, I'll leave to others.

[cj]

TYPE SCR WINS WIN% AVWINPRICE
ALL 113,698 13879 12.0 12.53
VET 37594 4453 11.8 12.67
STEW 21932 2594 11.8 12.89
TRNR 22987 3093 13.9 11.40
AL_EL 16540 1790 10.8 14.69
OFFT 6490 775 11.9 12.07
MaTO 4271 684 16 10.67
XENT 3841 483 12.5 11.60
GATE 41 6 14.6 25.83

[cj]

And, with ROI...

TYPE SCR WINS WIN% AVGWP ROI
ALL 113,698 13879 12.0 12.53 1.50
VET 37594 4453 11.8 12.67 1.50
STEW 21932 2594 11.8 12.89 1.52
TRNR 22987 3093 13.9 11.40 1.58
AL_EL 16540 1790 10.8 14.69 1.59
OFFT 6490 775 11.9 12.07 1.44
MaTO 4271 684 16 10.67 1.71
XENT 3841 483 12.5 11.60 1.45
GATE 41 6 14.6 25.83 3.77

[Saratoga_Mike]

Great stats CJ - thank you.

[cj]

Great stats CJ - thank you.

Thank Stuball, I just did the formatting to make it cleaner. ROI is per $2 bet.

[therussmeister]

I presume gate scratched are mostly vet scratches.

Oh, and based on the data given I consider scratches to be neutral, not a negative. They win about the same rate as randomly chosen horses.

[thaskalos]

So...where did this 5% figure come from?

[cj]

So...where did this 5% figure come from?

Like many things in this game, it is a myth.

[raybo]

So...where did this 5% figure come from?

This quote ought to give you an idea of where it came from.

I've been reading the "Has Anybody Here Heard Of…." thread and it mentioned a 2008-ish or so interview with Jim Clark, Director of (the now defunct) In-Form Handicapping System. In that interview, Clark makes the following claim:

"The percentage of horses winning off a scratch…is less than 5%, so you can pretty well toss them." :rolleyes:

[cj]

Another myth is that betting every horse will cause somebody to lose only the takeout, which would be an ROI of around 1.74 per $2 bet if the takeout averages at 18%. (I don't know the actual number)

But, in reality, it isn't close to that. Betting $2 on every horse the runs will result in lost 25% of your cash, not 18%.

[DeltaLover]

Like many things in this game, it is a myth.

:ThmbUp:

[DeltaLover]

Another myth is that betting every horse will cause somebody to lose only the takeout, which would be an ROI of around 1.74 per $2 bet if the takeout averages at 18%. (I don't know the actual number)

But, in reality, it isn't close to that. Betting $2 on every horse the runs will result in lost 25% of your cash, not 18%.

:ThmbUp:
100% Correct and one of the reasons why this game is (at least in theory) beatable

[davew]

Another myth is that betting every horse will cause somebody to lose only the takeout, which would be an ROI of around 1.74 per $2 bet if the takeout averages at 18%. (I don't know the actual number)

But, in reality, it isn't close to that. Betting $2 on every horse the runs will result in lost 25% of your cash, not 18%.

breakage is a bitch, especially with the favorites

[cj]

Another myth is that betting every horse will cause somebody to lose only the takeout, which would be an ROI of around 1.74 per $2 bet if the takeout averages at 18%. (I don't know the actual number)

But, in reality, it isn't close to that. Betting $2 on every horse the runs will result in lost 25% of your cash, not 18%.

Wow, I have to stop posting from my phone. That is horrible English!

[cj]

breakage is a bitch, especially with the favorites

I think breakage only accounts for about 1% of that. If you bet favorites only, you would actually substantially lower your losses from 25% to about 17%. I imagine betting one horse at random would fall somewhere in the middle, though closer to 17 than 25.

[appistappis]

Like many things in this game, it is a myth.


But if its packaged nicely, you can say anything and get $39.95 for it.

[cj]

But if its packaged nicely, you can say anything and get $39.95 for it.

What is selling for $39.95? I think those days are gone, for the most part. People have wised up to the shills.

[appistappis]

most have, thank god, but there are still some around in the back of the form and what not.

[PaceAdvantage]

most have, thank god, Why thank god? Wouldn't YOU (being the sophisticated player that you are), WANT PEOPLE to be buying these worthless systems and handing you their money at the window?

[appistappis]

who said I was sophisticated??

[Krudler]

Thanks Stuball for the stats and CJ for the formatting and ROIs. As has been previously noted, it appears that the 5% winners after a scratch stat doesn't hold up in today's game. That answers the question I had when starting the thread.

I'm still curious, however, to hear Jim's thoughts on the stats.

So...Jim, what's your take on Stuball's data? Might you interpret it differently? Do you have stats that contradict Stuball's seeing as you stood behind your theory in a recent post? Look forward to hearing from you.

[DeltaLover]

Bo doubt, the 5% after a scratch only exists in the creative imagination of the author. This should be clear by now.

More than this, I would like to know if there exist ANY OTHER FACTOR THAT CAN BE DESCIBED USING PAST PERFORMANCE DATA ONLY (NOT INCLUDING MORNING LINE OR CROWD ODDS) having such a low winning percent.

Also what is the minimum winning percentage you can guarantee even for the most ugly looking horse?

[PaceAdvantage]

who said I was sophisticated??Point taken.

[Jim Clark]

Folks, I see that my mention of the "scratch board" created quite a debate and while I have not been keeping a % log recently I would challenge that the numbers are higher than 5%. Case in point yesterday for the three tracks I looked at: Churchill, Hawthorne and Woodbine. 22 scratches of note with 1 winner.

I was playing Woodbine last night and note that Hockey was 3/2 and a double scratch in between races. Yes, I through him out and landed on the winner and exactor.

RACE 2

HORSE DATE RACE CLASS REASON
Hockey 16Nov13 10WO Md 25000 Trainer
Hockey 8Nov13 10WO Md 16000 Stewards

As a successful handicapper, my point is that handicappers NOT reviewing and analyzing the scratch board as part of their process are, in my opinion, missing a key part of the process. Play as you will but take note yourself.

[Longshot6977]

Bo doubt, the 5% after a scratch only exists in the creative imagination of the author. This should be clear by now.

More than this, I would like to know if there exist ANY OTHER FACTOR THAT CAN BE DESCIBED USING PAST PERFORMANCE DATA ONLY (NOT INCLUDING MORNING LINE OR CROWD ODDS) having such a low winning percent.

Also what is the minimum winning percentage you can guarantee even for the most ugly looking horse?

Not trying to be cute, but simply tossing a horse coming off a winning maiden claiming race might fit your scenario using just PP's. Unless the horse won by 6 or more lengths, it is a throwout. The % of winners is very likely low on a repeat winner off a winning MCL race. It's usually the fav in its next race, so look elswwhere for value. Any databasers with info on this?

[DeltaLover]

Not trying to be cute, but simply tossing a horse coming off a winning maiden claiming race might fit your scenario using just PP's. Unless the horse won by 6 or more lengths, it is a throwout. The % of winners is very likely low on a repeat winner off a winning MCL race. It's usually the fav in its next race, so look elswwhere for value. Any databasers with info on this?


What you say here, represents a very good example of a typical fallacy that the empirical player falls victim of very often. He tends to combine what seems rational to him with some selective data that he can remember, reaching a completely wrong conclusion.

The angle you are suggesting here described as:

Querying my data base for horses running right after their maiden claimer win which was achieved with less than 6 lengths, I am getting the following results:

count: 3312 winners: 392.0 win%: 11.84 ROI: 0.84:

Which is completely opposite to what you where anticipating. This is why we need hard data to support any of our claims and this is also a reason why the vast majority or related literature is completely unsupported by data, just representing the opinion of the writter.

For the history, the exact same universe of races showed the following results for horses coming of a maiden claimer win over 6 lengths:

count: 561 winners: 90.0 win%: 16.04 ROI: 0.64

Not considering winning lengths we get the following:

count: 3873 winners: 482.0 win%: 12.45 ROI: 0.81

I will repeat that it is extremely unlikely to find a simple handicapping factor showing a winning percent as low as 5% by itself. Any writer who claims to have find such a factor, he better present us the data behind his claim, otherwise he is still a novice handicapper.

[davew]

Folks, I see that my mention of the "scratch board" created quite a debate and while I have not been keeping a % log recently I would challenge that the numbers are higher than 5%. Case in point yesterday for the three tracks I looked at: Churchill, Hawthorne and Woodbine. 22 scratches of note with 1 winner.

I was playing Woodbine last night and note that Hockey was 3/2 and a double scratch in between races. Yes, I through him out and landed on the winner and exactor.

RACE 2

HORSE DATE RACE CLASS REASON
Hockey 16Nov13 10WO Md 25000 Trainer
Hockey 8Nov13 10WO Md 16000 Stewards

As a successful handicapper, my point is that handicappers NOT reviewing and analyzing the scratch board as part of their process are, in my opinion, missing a key part of the process. Play as you will but take note yourself.

could you tell us what you put on your scratch board?

[Robert Goren]

What you say here, represents a very good example of a typical fallacy that the empirical player falls victim of very often. He tends to combine what seems rational to him with some selective data that he can remember, reaching a completely wrong conclusion.

The angle you are suggesting here described as:

Querying my data base for horses running right after their maiden claimer win which was achieved with less than 6 lengths, I am getting the following results:

count: 3312 winners: 392.0 win%: 11.84 ROI: 0.84:

Which is completely opposite to what you where anticipating. This is why we need hard data to support any of our claims and this is also a reason why the vast majority or related literature is completely unsupported by data, just representing the opinion of the writter.

For the history, the exact same universe of races showed the following results for horses coming of a maiden claimer win over 6 lengths:

count: 561 winners: 90.0 win%: 16.04 ROI: 0.64

Not considering winning lengths we get the following:

count: 3873 winners: 482.0 win%: 12.45 ROI: 0.81

I will repeat that it is extremely unlikely to find a simple handicapping factor showing a winning percent as low as 5% by itself. Any writer who claims to have find such a factor, he better present us the data behind his claim, otherwise he is still a novice handicapper.Try a horse who has been off at least 2 years.

[DeltaLover]

Try a horse who has been off at least 2 years.

This angle represents a very insignificant factor, since out of 155,789 starters only 100 appear to return after a layoff of more than 700 days while six of them managed to win.

total_starters 155,789 count: 100 winners: 6.0 win%: 6.00 ROI: 0.34:

These type of factors present almost zero betting value.

[classhandicapper]

What you say here, represents a very good example of a typical fallacy that the empirical player falls victim of very often. He tends to combine what seems rational to him with some selective data that he can remember, reaching a completely wrong conclusion.

The angle you are suggesting here described as:

Querying my data base for horses running right after their maiden claimer win which was achieved with less than 6 lengths, I am getting the following results:

count: 3312 winners: 392.0 win%: 11.84 ROI: 0.84:

Which is completely opposite to what you where anticipating. This is why we need hard data to support any of our claims and this is also a reason why the vast majority or related literature is completely unsupported by data, just representing the opinion of the writter.

For the history, the exact same universe of races showed the following results for horses coming of a maiden claimer win over 6 lengths:

count: 561 winners: 90.0 win%: 16.04 ROI: 0.64

Not considering winning lengths we get the following:

count: 3873 winners: 482.0 win%: 12.45 ROI: 0.81

I will repeat that it is extremely unlikely to find a simple handicapping factor showing a winning percent as low as 5% by itself. Any writer who claims to have find such a factor, he better present us the data behind his claim, otherwise he is still a novice handicapper.

These days a lot of maiden claiming winners go into NW2 lifetime claiming races after they graduate. So they are often either not rising much in class or in rare cases actually dropping in class. The stats might be different if it was limited to maiden claiming winners going into open claimers of a similar price. Then it would be a legitimate jump in class in most cases.

[Horseplayersbet.com]

I think there is more too vet scratches than meets the eye. A horse off a vet scratch who is off over 21 days probably performs worse than one on average than one off less than 20 days. Vet scratched horses who have a vet scratch since their last race, with a layoff of over 30 or 35 days probably under perform on average too.

Vet scratches may not matter as much for high percentage trainers and most likely matter less the higher the claiming price.

I don't do the data base thing, so I might be falling for myths myself, but I think average figures decrease by 2-3 lengths if a horse is off a vet scratch and a 21 day plus layoff. Of course, I could be wrong.

An interesting query would be trainer (under 13% win percentage), 21-60 day layoff, vet scratch since last race, claiming price $16,000 or less.

Another good query would be the race after a horse who finished 1st or 2nd off a vet scratch. Intuitively, I expect a major bounce in many cases.

[Fingal]

That hasn't been my experience.

In So Cal horses are sometimes entered two days in a row.
Depending on the card, the trainer can scratch from day 1 and stay in day 2.
I haven't kept stats, but 5 % sounds low.

It depends on the type of race scratched out of. Say a trainer had an upcoming 3 yr old, entered in a stake, scratched & then the next day the horse runs in the ALW conditions they're still eligible for.

Then for a nomination fee the trainer got to see how that race filled, then decided if he or she wanted to scratch to go in the easier spot.

[Longshot6977]

What you say here, represents a very good example of a typical fallacy that the empirical player falls victim of very often. He tends to combine what seems rational to him with some selective data that he can remember, reaching a completely wrong conclusion.

The angle you are suggesting here described as:

Querying my data base for horses running right after their maiden claimer win which was achieved with less than 6 lengths, I am getting the following results:

count: 3312 winners: 392.0 win%: 11.84 ROI: 0.84:

Which is completely opposite to what you where anticipating. This is why we need hard data to support any of our claims and this is also a reason why the vast majority or related literature is completely unsupported by data, just representing the opinion of the writter.

For the history, the exact same universe of races showed the following results for horses coming of a maiden claimer win over 6 lengths:

count: 561 winners: 90.0 win%: 16.04 ROI: 0.64

Not considering winning lengths we get the following:

count: 3873 winners: 482.0 win%: 12.45 ROI: 0.81

I will repeat that it is extremely unlikely to find a simple handicapping factor showing a winning percent as low as 5% by itself. Any writer who claims to have find such a factor, he better present us the data behind his claim, otherwise he is still a novice handicapper.

Thanks for the data DL. However, I feel if a horse wins 11.84%, (which is still low, but not as low as the 5% you were asking about) he loses about 89% of the time. It's usually those times we are usually making our bets on them. I'd rather stay away from a high losing % situation. You need to bet every race/entry to be helped by the 11% stat. I don't know anyone who bets every single race/entry for a particular scenario. Plus the odds are usually lower on a repeat MCL winner by over 6 lengths.

But I'm sure someone will find a factor that hits about 5% if they try hard enough; like horses coming off a 4 race Beyer increase or grey horses on a Monday in post 9 or something silly. Anyway, thanks for the stats. :)

[cj]

Thanks for the data DL. However, I feel if a horse wins 11.84%, (which is still low, but not as low as the 5% you were asking about) he loses about 89% of the time. It's usually those times we are usually making our bets on them. I'd rather stay away from a high losing % situation. You need to bet every race/entry to be helped by the 11% stat. I don't know anyone who bets every single race/entry for a particular scenario. Plus the odds are usually lower on a repeat MCL winner by over 6 lengths.

But I'm sure someone will find a factor that hits about 5% if they try hard enough; like horses coming off a 4 race Beyer increase or grey horses on a Monday in post 9 or something silly. Anyway, thanks for the stats. :)

Problem with this is that 11.84% probably applies to tons of categories of horses, including any random horse winning a race.

[ezpace]

TRAINER SCRATCHES once or twice
then drops in class l call it shoppin an droppin
certain trainers . DYODD