Beyer figs

[cj]

Originally Posted by Afleet

that a good post

The three year best is not a good measuring stick. It is a little better than the track record maybe. How many 9f races do people think are run on dirt at these tracks in a three year span?

[dilanesp]

At some point, people need to stop obsessing about raw time.

Here's a vivid example.

The slowest Hollywood Gold Cup in history, going all the way back to 1938, was run by Game on Dude (2:04 flat) in 2012.

The fastest Santa Anita Handicap in history, going all the way back to 1935, was run by...

Game on Dude (1:58 flat) in 2014.

I assure you that Game on Dude did not get 6 seconds faster between age 5 and 7.

The 2012 race DID earn a lower speed figure-- not 6 seconds lower, but lower. But it was also run over a much slower synthetic surface, whereas the 2014 was run over a Santa Anita dirt strip that was one of the fastest in history.

There are a lot of disagreements about particular methodologies about converting raw times into usable numbers. Those disagreements are interesting. But-- YOU MUST AT LEAST USE ONE OF THEM. Track speeds vary, a LOT, and you have to use some valid methodology for adjusting times based on track speed before making handicapping conclusions based on them.

And you can't simply dismiss Justify as a slow horse when he's running his times over a surface that is quite a bit slower than Santa Anita was just a couple of years ago.

[CincyHorseplayer]

I always use the 6% margin of error rule when I look at a race. My rule so...If the top figure is X then you accept as competitive in the race any figure within the 94 percentile. I find this rule to be at worst around 80% and normally around 87%. We keep hearing about Buckpasser X and RAN lines being the esoteric realities of this game and since 1990 RAN is only at 58%. Pedigrees change too. It may never be that good again in the next 18 years. But fast will always be good. It will always outperform nearly all other statistics. Top figure is not only what figure players look at. That perversion of thinking being applied by non figure players is as bad as pace players betting only frontrunners! I understand what is going on in pedigree. And as applied to the Derby it's not that complicated. The pedigree work I do with turf races take some real artful insight. Seeing how it is used here and the scorn by the same players against figures reminds of of someone trying to hit a hard ball with a wiffle ball bat. Looking for the next Giacomo to validate their entire being as their entire playing life revolves around this single race. Talk about outliers, stupidity, and BS! We've got a whole legion of players who embody all those virtues and they all try to preach to us this time of year. Cheers!

[Afleet]

Originally Posted by cj

The three year best is not a good measuring stick. It is a little better than the track record maybe. How many 9f races do people think are run on dirt at these tracks in a three year span?

not many which is sad, but the 6F, 1 1/16 would be relevant

[cj]

Originally Posted by Afleet

not many which is sad, but the 6F, 1 1/16 would be relevant

Better but still not great. And all the preps shown in the example were at 9f.

[CincyHorseplayer]

Originally Posted by cj

Better but still not great. And all the preps shown in the example were at 9f.

I know I'm going out of the way here but you talking about distances reminded me of this. How do you feel about the route distances for G1 races? I would like to see them get a little longer again. I think there should be very few G1 races run at 9f. They should be 10f. The mile is a specialty distance worldwide, IMO there is no reason for other races that are graded to be so hugged up to it. Much like how Belmont has run more turf races at 10-11 furlongs I'd like to see more dirt races run further too. It would long term improve the breed IMO. Where you at on this?

[Gerard02]

I agree. More races should be longer distances.

[Robert Fischer]

Was OBVIOUS to me watching live, that either the track was playing slow, or the fractions were malfunctioning/mis-timed.

Wasn't a suicidal pace, but it was a moderate-to-fast pace. There was a slight danger that Mike Smith was going to blow it by going too fast.
He didn't reach that point, but he flirted with it.

I'm watching the race with a bet down, and I'm saying "Easy, Mike... don't go too fast now... take it easy!... Why is he being so aggressive!?".

Believe it or not, I watch a lot of races.

Doesn't matter to me what numbers show up on the screen or whether they show up or not. I can kind of tell what's going on.


I think the figure makers did a great job. They should have. The card and the recent racing has shown the track has played slower. It's not like the track supposedly 'changed' between races xyz because they dragged the tractor or whatever...

[Wiley]

Originally Posted by Robert Fischer

Was OBVIOUS to me watching live, that either the track was playing slow, or the fractions were malfunctioning/mis-timed.

Wasn't a suicidal pace, but it was a moderate-to-fast pace. There was a slight danger that Mike Smith was going to blow it by going too fast.
He didn't reach that point, but he flirted with it.

I thought the same thing, why is Smith urging Justify on with a clear and easy lead early in the race. I assume Baffert told him to make him work some to get as much out of the race as possible, and maybe also see what he will have for a finish. Pretty bold move considering he had to finish one or two to make the Derby, or a ton of confidence in the horse.

[depalma113]

Until someone can logically explain why Bolt d'Oro is not credited with a 113 Beyer in the Frontrunner, I am not buying any Beyer speed figure coming out of Santa Anita.

If they can arbitrarily drop a horse 13 points, they can certainly raise it too.

[f2tornado]

Originally Posted by Wiley

I thought the same thing, why is Smith urging Justify on with a clear and easy lead early in the race...

Adjusting the final clock by 2.2 (or just under a quarter second per panel) to match the time California Chrome won at (with support by equal Beyer and similar Equibase figures) yields a 3/4 fraction in 1:11.14 which is on par with the other preps short the Ark.

I don't recall a Derby with so many horses that finished strongly in final prep. I really think a horse will need to be within about five lengths of the leader at the 3/4 pole to pull off a victory. Ten horses in the gate ran a final 3/8th in 37.8 or less, none dramatically faster than the others. I'd normally not get too caught up in post draw but it is seemingly more critical this year given much of the field has the exact same style. Justify backers should hope he draws inside and placed near the front then it's just a matter of staying power through that final panel.

[Robert Fischer]

Originally Posted by depalma113

Until someone can logically explain why Bolt d'Oro is not credited with a 113 Beyer in the Frontrunner, I am not buying any Beyer speed figure coming out of Santa Anita.

If they can arbitrarily drop a horse 13 points, they can certainly raise it too.

I don't know how to make a Beyer.

I would be interested if you laid it out.


One of the blessing/curses in Beyers is that they have seem to have a projection element, like a class element.

Bolt got a 103 and Solomini got a 90.

Both of those figures appear to 'fit' in hindsight. Doesn't mean that they were correct, just that they look sensible in hindsight.

If Bolt gets a 113 instead of a 103, does that mean Solomini should have earned a 100 instead of a 90? How does the scale for 2nd work if you change the winner's number?

I don't have any problem saying the Frontrunner was Bolt's fastest race.

I have it "3rd start G1 Front Runner S. 1 1/16M - Sharp effort. Broke with the pace, settled nicely just behind and outside the leader, pulled away from the field at the end of the far turn. Good energy in the stretch.".

I didn't really think it was a '113' level performance (today that basically means "WOW! this horse absolutely FREAKED!"), but if they gave him a 105 or 106 instead of the 103, I wouldn't have questioned the legitimacy.

[dilanesp]

Originally Posted by Robert Fischer

I don't know how to make a Beyer.

I would be interested if you laid it out.


One of the blessing/curses in Beyers is that they have seem to have a projection element, like a class element.

Bolt got a 103 and Solomini got a 90.

Both of those figures appear to 'fit' in hindsight. Doesn't mean that they were correct, just that they look sensible in hindsight.

If Bolt gets a 113 instead of a 103, does that mean Solomini should have earned a 100 instead of a 90? How does the scale for 2nd work if you change the winner's number?

I don't have any problem saying the Frontrunner was Bolt's fastest race.

I have it "3rd start G1 Front Runner S. 1 1/16M - Sharp effort. Broke with the pace, settled nicely just behind and outside the leader, pulled away from the field at the end of the far turn. Good energy in the stretch.".

I didn't really think it was a '113' level performance (today that basically means "WOW! this horse absolutely FREAKED!"), but if they gave him a 105 or 106 instead of the 103, I wouldn't have questioned the legitimacy.

I am going on memory, but here are the steps I followed when I made speed figures as a teenager:

1. Start with a commonly run distance and class level and collect a large sample of data. For instance, at Santa Anita these days, you might use allowance optional claimer nonwinners of 2 at 6 furlongs. (Back in the 1980's I used $12,500 open claimers at 6 furlongs, but now there aren't so many of those.)

2. Take the mean of all times at that distance and class level and assign it an arbitrary number such as 70 or 80 or 90.

3. Now collect another large sample of data for the same class level at other distances. 5 1/2 furlongs, 6 1/2 furlongs, 7 furlongs, 1 mile, 1 1/16 miles. Each par time will get the same number that you arbitrarily assigned.

4. Now, at 6 furlongs, assign a point value for each 1/5 of a second (or nowadays, 0.20 seconds) interval. Such as 2 points.

5. That same point interval should be a different amount of time at the other distances, because the horses are traveling faster at the end of shorter races and slower at the end of longer races. So divide the .20 seconds by the par time of the 6 furlong race, expressed in seconds. Then multiply it by the par time for each other distance, and you will get the equivalent number of seconds for each distance. Based on that, you can construct a beaten time chart for each distance. Each 2 points will be .20 seconds at 6 furlongs, it will be a little less than .20 seconds at 5 1/2 furlongs, and it will be significantly more than .20 seconds at 1 1/16 miles, for instance.

6. Using the beaten time chart, you can construct a speed figure chart for each distance that assigns each time a number, by, for instance, going up and down 2 points for each .20 at 6 furlongs, and using the other numbers you just calculated with the other numbers. Those are your base speed figures.

7. Go back and take average base speed figures over large samples at various common class levels at the track. These are your par times. After you have a large number of par times, use those times to project the speed figures for races going forward. Compare the actual numbers run by winners to the par times, and average the differences out to create an estimate of whether the track is faster or slower than normal. This is your track variant. Add or subtract the track variant to your base numbers. These will be your preliminary speed figures.

8. Once you have a few weeks of speed figures calculated this way, instead of continuing to use the par times, start using the actual numbers run by horses in previous races to project what they will run in each race. Once again, compare the projected numbers to what they actually run. Use a par time when you don't have enough data. Once again, average these out each day to create an estimate of track speed, which is your track variant. Add or subtract the track variant to your base numbers. Now you are calculating real speed figures.

[Gerard02]

Ill need to take a Motrin and read that again.

[GMB@BP]

Originally Posted by Robert Fischer

I don't know how to make a Beyer.

I would be interested if you laid it out.


.

https://www.youtube.com/watch?v=aketcA_hkoM