28 June 2009

The Perfect Past Performances Just Do Not Exist

Life for the handicapper can be made so much more simpler.

Today's horseplayer is likely to play more than one track a day. Speed handicapping is required in many instances, especially for the many who don't rely on computer print outs in order to determine their bets.

We've seen tremendous improvements in the past performances that were available in the early 70's to today. Speed figures, pace figures, turf stats, workout ratings, trainer stats, jockey stats, etc. have all helped level the playing field.

Still, there are many holes and unanswered questions in the vast array of past performances that are available to the consumer right now. These holes can be filled in by doing extra research: watching race videos and doing your own track bias ratings, and marking down wind speed daily. But I think these things should be covered by those who chart the races for a living, and they should be available in the past performances.

Here is what is needed to make the perfect past performances:

If you want to have some fun, go search the internet for the official amount of feet in a length. You'll get answers anywhere from 7.5 to 10 feet. Horses range in size too. So when a tiny horse wins by his length, does he win by a length, or really 3 quarters of a length?

In other words, lengths are subjective. Not only that, but I've tried to get a straight answer from actual chart callers and other experts with regard to the following circumstance and I've yet to find a consistent answer, or one that I can even understand:

The #2 horse has a 12 (subjective) length lead when he crosses the wire. The #3 horse was second when the #2 crossed the wire, but was tiring very quickly. The #5 horse who was 16 (subjective) lengths from the #2 when he crossed the wire first, wound up plodding along to finish second by 2 (subjective) lengths. Oh, and it was a muddy track, and the second horse finish the race exactly 4 seconds after the winner.

How many lengths does the chart show the #2 won by? 12 lengths? 16 lengths? or 20 lengths? And what about the third finishing horse, did he lose by 14 lengths, 15 lengths or 22 lengths, or 23 lengths (as his final time wound up to by 4.60 seconds slower than the winner's time)?

I wish I knew the answer. And I don't think anyone does definitively, and what may be true at one track, may not be true at another.

Horses do go slower the farther they go, and then you have to factor in the speed of today's track and the wind speed as well when it comes to how fast a horse is going at the end of the race.

When a horse loses by 10 lengths going 5 furlongs, does that mean he ran 2 seconds slower than the winner, the same as when a horse loses by 10 lengths going a mile and an eighth? Even though they are subjective lengths, it takes a lot longer for a horse to cover 10 lengths go further. And what if the horse is really tiring or coming on strong?

I want to know the final time of every horse so I can determine the lengths lost myself, and since I'm a speed handicapper, it would be nice to know the exact times I'm dealing with.

I can live with guesstimations when it comes to lengths beaten at every other call, but not the finish.


This is a no brainer. The chart maker simply jots down the wind speed every race. And that wind speed should show up in the past performances. They somehow have the technology to do this for quarter horses, so I know it exists.

This is important because when the wind gets up over 20 mph in a race, it definitely can create a bias. Horses with the wind at their back in the stretch have an advantage if they are closing and have more energy in reserve, while horses on the lead have an advantage if entering the stretch against a strong headwind.

A simple notation like 25b or 25h means that it is either a backstretch wind or a homestretch wind of 25 mph.


Steve Klein, in his book The Power Of Early Speed came up with a universal way to quickly come up with a relative speed bias number. It isn't rocket science.

Without giving it totally away, it has to do with the first call positions of most winners on a daily card, and then averaging them out, giving certain weights depending on the starting position.

I do think it can be done more effectively by incorporating the top two finishers and the favorite, but the principles will remain the same, and it could be calculated by a computer program in a nanosecond (or whatever the term is for almost no time).

The number can be scaled to be out of 100, where the higher the number, the more likely a speed horse lasted.

Wouldn't it be nice if one could look at the past performances and see that a horse was in the back of the field on a day that had a 72 speed bias rating?


The same way a speed bias can be quickly be determined, a rail/outside bias can also be determined.

Use the first two finishers and the favorite (if not in the first two finishers) and plot where they were most of the race. Anything 2.5 path or greater is consider outside, and anything less than 2.5 path is considered rail/inside.

Again, an average base on the results of the day is made, and a number out of 100 is placed on the running line of each horse's past performances. A 22, for example, would mean a good outside, bad rail, while a 60 would mean a good rail, and a bad outside bias.


The chart called needs to put down how wide each horse was at three different points of the race. For one turn races, average backstretch, average turn, and average stretch. For races over a mile, average first turn, average backstretch and average final turn.

In a sprint 3R2 means that the horse was 3 wide on average down the backstretch or midpoint of the backstretch, on the rail most of the time on the far turn, and 2 wide either mid stretch or on average during the stretch.

Checked or boxed in (like you see in harness racing) can be illustrated with a small c or b after the specific rating. Using my above example, if the horse was checked in the far turn, the line would read 3Rb2.

This data can replace the comment on the horse's line if space is needed to be accommodated for.


The harness racing folks are way ahead of the thoroughbred folks when it comes to showing the class and competition a horse could have run against if not scratched.

Sure, many past performances available show some scratch info at the end of the card, but it means flipping back and forth, and the information of who the horse would have run against is only available if you dig up your old past performances, or you have a good memory, or happen to have a one-time good memory.

Just like when a horse was previously trained by a someone gives the player something viable to consider, so would seeing a "vet scratch" line 3 weeks ago, with a bad workout following it, for example.


Simply put the jockey's name in italics if he or she was not the named program rider.

Again, the more info, the better for the bettor.

Surely the good people at Equibase can get the show on the road, and make it so that the PERFECT PAST PERFORMANCES can at least be available.

One more thing, if you are a bettor who wants a chance at more preferential treatment by the race tracks, please click here to join HANA, it will take you less than a minute, and it is completely free. We need the numbers to have more clout.


Amateurcapper said...

How much do you suppose that would cost?

I think that would be too labor-intensive to mass produce at a reasonable rate.

Anyway, at least I still have an excuse when I lose :-)!

Anonymous said...

TRakus web site has actual times for each horse at every pole

Anonymous said...

Some of these ideas wouldn't be so hard to add. Final times on each horse when they all cross the same teletimer, why not? Wind direction, it's on the TV screens, why not note it? Change of jock, how hard to add italics? This is useful for claiming too.