Calucating %E -- By Hand

[andicap]

I thought I knew the formula for computing %E, but when I ran a race through my old MPH program's profile it came out differently. Who's right, me or MPH?

1/17 AQI first race. 6f.

in fifths (MPH is old can't do hundreths)
48 114.1
MPH does two calulations, the race winner and the pace of race.
It gave the %E of the pace of race (48 and 114.2 in hundreths) as 52.37

I calculated it as 52.19

4f is 2640 feet/48 = 55
FF is 1320 feet/26.2 (in hundreths) = 50.38

55/105.38 = 52.19

who's right?

[Dick Schmidt]

You are correct. Either MPH has the formula slightly wrong or Tom Brohammer was fooling around with %E again when it was written. You are using the correct formula originally developed by Doc Sartin.


Dick


If you can't beat your computer at chess, try kickboxing.

[Kreed]

One if by hand, two, by sea. guys, stop being RetrO & INSIST that ALL our
race venues get the HongKong gps method of timing winners & losers. Stop
with the 8-9-10 FEET controversy (LOL LOL) .... all ridiculous measures.
if u (WE) had better data, maybe Formula would sing Punk.

[delayjf]

Interesting that I just read in an old Follow Up were Doc was advocating using median energy as opposed to %E. He claimed that it gave a more accurate picture of the energy expended. Median energy is calculated as follows:

4 furlong time / final time so in the example cited,

48 / 74.2 = 64.69%

Not sure if this improved the usage of %E or not.

[keilan]

Jeff

Energy is only as good as the person using it :)

[46zilzal]

4 furlong time / final time so in the example cited,

48 / 74.2 = 64.69% .

Another method: Sum the velocity of all three fractions.....Add up the velocity of the first two fractions. Sum 1+2/ sum of all three and you get the same thing % median....a great factor to show horses going off form....false favorites often are ELIMINATED this way.

[46zilzal]

I
1/17 AQI first race. 6f.

in fifths (MPH is old can't do hundreths)
48 114.1
MPH does two calulations, the race winner and the pace of race.
It gave the %E of the pace of race (48 and 114.2 in hundreths) as 52.37

I calculated it as 52.19

4f is 2640 feet/48 = 55
FF is 1320 feet/26.2 (in hundreths) = 50.38

55/105.38 = 52.191

Lines going IN: 70 is too high to win much there and the 10 won it
% median e/l
#2 Half line 1 71.0 15.9
#3 Triple line 3 70.4 16.3
4 Deputy line 2 68.7 2.1
5 Afflue2 69.3 7.5
6 Unbridl2 68.3 -0.3
7 Deeli2 70.7 18.5
10 Sky B1 68.1 -5.2
11 Golde1 68.5` -1.4
a better way to follow a track is the early/late balance which is calcualted using this, but I do not know the exact calculation...Keep track at each distacne and you can eliminate several each race...You need to know what they are PROJECTING into this race

[andicap]

I wanted to know this in order to look calculate at %E levels from races in the charts to get an idea of the differences between classes/sexes/ages at different tracks. You can also tell by looking at the pace of race %E levels if a particular race had an unusually fast or slow pace.

On Dec. 30 at AQI, the %E pace of race varied from 50.17 to 51.73. I would wager that the front-runners from the 50.17 race will not do as well going foward but the ones in the 51.73 race are eligible to show better with a softer pace next out. I would upgrade closers from 50.17 race tho.

Using MPH made it somewhat faster, but it appears incorrect. (MPH also calcuates median energy, but I didn't check whether the software was correct there or not because nothing I uses automatically gives me %M in the PPs just %E.)
With HTR I have also used a variation of Sartin's % early where you compare the first fraction against all three using an HTR velocity based rating that is NOT adjusted for distance. (Keilan keeps telling me to calucate energy before adjusting figs for distance). I try to compare each horse's level using pacelines that would generally reflect what to expect today in terms of pace.


In the past when I used energy in a totally different fashion with Thorovision software (now defunct), I had real success using a figure derived from each horse's PPs to determine a model rather than going over the charts and calcuating the %E as you would a track profile.

But this was a wholly different method whereby horses' early energy expenditures were compared to each other on a 0.0-8.0 scale (early being lowest) that was dynamite as long as you only used races -- up to 3 "good" ones -- that reflected a realistic pace.

I would get models at some tracks -- usually short-term ones -- that were pretty tight and, baring some weird pace scenario, the winner had to come from within that range. I recall one AQI period where virtually every horse had to have a 3.0-5.5 in order to win. A real presser's track. Longshots came in by the bucketful because favorites out of the range would lose. Often there was no "bias" -- horses of all types would win based on that race's pace match-up -- but when one existed, watch-out!

The guy who helped design the method e-mailed me the caluclations but I have never been able to duplicate it from his notes and he never responded to my follow-up. You had to convert the PLs used into velocity, convert to beaten lengths (they used 10 feet a length), and derived a scale from that somehow.

I wanted to use Thorovision just for that figure alone but the program had Y2K issues, and the programmer died a few years ago so it is basically defunct.

For those familar with the program it was the "DA" figure in the energy screen.

[Tom]

Jeff

Energy is only as good as the person using it :)

Thanks for reminding me! :rolleyes:

[46zilzal]

Have never found it advantageous to keep % early or % median by CLASS (artificial, manmade anyway) just by DISTANCE and track. Always note that it is HIGHER naturally in YOUNGER animals (they do not apportion their runs like older, mature horses). Only thing it really helps one with, is telling how much of a bias is there WITHOUT referring to the VISUAL perception of the races (which is often very flawed)...Many days in N.Y. entire cards are run on the front end only (when it is very cold) and knowing this, PRE- race, is a great boon on wagering.

[JimL]

Jeff

Energy is only as good as the person using it :)I

I know this Keilan, that is why I keep waiting for you to work another race for us! Before or after the race, I dont care JimL

[46zilzal]

going off form : use the % median as it goes up over a three race pattern WIHTOUT improvement in total energy and you've found a false favorite

[46zilzal]

% med e/l
1 Marqu1 67.4 -9.7
1 Marq3 69.9 12.3 Third, other half of th e/l dominance
3 Search2 68.6 2.0
5 Alic1 69.5 7.1
5 ALic3 69.1 5.8
6 Allen1 67.5 -5.5
6 Allen2 69.0 5.7
7 Fly W2 69.5 10.9 WINNER one of the top two E/l's


Works over and OVER

[keilan]

going off form : use the % median as it goes up over a three race pattern WIHTOUT improvement in total energy and you've found a false favorite

Caveat --- if the pace is faster in the last 3 races than previous races all this tells you is the horse is probably a class level too high. Further to that if the horse's energy number is increasing I'd suspect he isn't winning too many of those 3 races and is unlikely to be the favourite.

Your previous post is on the money 2 of your 3 points.

1) Have never found it advantageous to keep % early or % median by CLASS (artificial, manmade anyway) just by DISTANCE and track.
2) Always note that it is HIGHER naturally in YOUNGER animals (they do not apportion their runs like older, mature horses).
3) Only thing it really helps one with, is telling how much of a bias is there WITHOUT referring to the VISUAL perception of the races

Without doing point #1 you'd have not known that point #2 is correct. There are also other things a player will learn from charting the winners for class, distance , sex etc

[46zilzal]

Don't look at the class structure unless it is to find damaged goods trying to GIVE away cheaper: HORSES run they don't kow WHO they are running against...If a pace is too fast for a horse, pattern changes and one can tell it did NOT do well......class strucuture is man made in the racing secretaries office NOT out on the race course.


TOOO MUCH in "read into" races this way

[46zilzal]

Really short sprints often tend to run "sustained" of all things. There is a natural tendency to "down play" lines from these ultra short sprints when streching out. That bias will cost you: Gatito Fuerte in the Aqu's third today coming off TWO 4.5f sprints won at $10.20 ....Last week's night cap (GP 1/22) tagged another one (Joe Peg) at long odds by NOT being biased by this ultra short sprint...

[andicap]

46.
In your example, what does the "late" number represent if the "early" figure is %M? The higher the number the more late energy? But how does one arrive at this figure?

And is total energy F1+f2+F3 velocity?

[46zilzal]

The early/late BALANCE is just another way of representing the %median. ZERO would mean a perfect balance of energy early and late, anything postive is on the early side(i.e. 12.0) and antying NEGATIVE on the late (i.e. -13.5). The way %median is calculated a LATE horse would be around 66%..

Total is F1+F2+F3 %median is F1+F2/F1+F2+F3

[46zilzal]

grabbing these from a 8.5 at Aqueduct (the e/l changes with distance)
Total F1 F2 F3 med% early/late
S/P 155.3 53.6 51.9 49.8 67.9 -14.4
late 157.6 52.3 52.7 52.6 66.6 -27.6
early 155.09 54.5 53.4 47.1 69.6 8.0
E/P 155.44 53.4 53.9 48.1 69.1 -0.5
SUS 158.38 54.5 52.6 51.2 67.6 -8.5