Compiling Speed figures.

[Outlander]

I always start out with a single allowance for a meeting, and in my table of data (example below for yesterday) I have columns that show the difference between the OR and my rating based on the allowance, I hope you can see it okay - I removed some columns (date for example) to try to squeeze it in without removing anything you'd need to see....

View attachment 81084

OR-Rtg is the first thing I glance over after compiling ratings, they should all be (ideally) showing about 30 (OR is 30 higher than rating), in reality the best three runs of the day will average this while the rest get steadily higher (ie running lesser figures than they might). Here we see nothing past 57, which is okay - no alarm bells. If the hurdles and chases were running differently I'd see perhaps all the hurdles between 25 and 45 say and the chases would maybe be at 55-80 or whatever.... ie I'm looking for the hurdles to be a different range to the chases, with one or the other obviously better than the other - you do have to accept that it won't usually be completely straightforward, you might find one race that is notably higher or lower than the others of its type.

If I think there might be an issue then I look at the s/f values, which shows the time fast/slow per furlong for the race. As you see above they are all reasonably similar with no obvious split in the range covered - Hurdles are 1.23, 1.53. 1.64, 1.45, 1.75 while the chases are 1.25 and 1.38, which fit happily into the hurdle range towards the lower end. You might see 1.2 -1.6 for hurdles and 2.3 - 2.5 for the chases (as an example), to go along with seeing both chases are showing bigger OR-rtg scores, that would be a pointer that the chases were running in slower ground.

Additional help comes from the 'This runner' column (ratings the winner has previously earned) that allows me to see if the rating for today is a reasonable fit with previous ratings (ie not 50 lbs better than before, or 50 lbs worse) and the 'spd rtgs' column that shows me ratings of winners of similar races at the track, which helps show if the rating is unusual for this class/type/track. These two columns can help confirm an opinion formed by looking at the OR-rtg and s/f, I wouldn't base a decision on this last pair if I hadn't already spotted a potential issue from the s/f and OR-rtg,.

If I decide there's a problem I then split the results data I've already processed from the bare results file and treat chases and hurdles as two separate meetings, and see what comes out, checking to see if it likes a better fit.

There is a large degree of personal opinion involved in this, it's deciding if 'it looks right' - I might reprocess things several times, thinking each rerun still doesn't feel right, until I hit on something I can live with.

Dave

I always start out with a single allowance for a meeting, and in my table of data (example below for yesterday) I have columns that show the difference between the OR and my rating based on the allowance, I hope you can see it okay - I removed some columns (date for example) to try to squeeze it in without removing anything you'd need to see....

View attachment 81084

OR-Rtg is the first thing I glance over after compiling ratings, they should all be (ideally) showing about 30 (OR is 30 higher than rating), in reality the best three runs of the day will average this while the rest get steadily higher (ie running lesser figures than they might). Here we see nothing past 57, which is okay - no alarm bells. If the hurdles and chases were running differently I'd see perhaps all the hurdles between 25 and 45 say and the chases would maybe be at 55-80 or whatever.... ie I'm looking for the hurdles to be a different range to the chases, with one or the other obviously better than the other - you do have to accept that it won't usually be completely straightforward, you might find one race that is notably higher or lower than the others of its type.

If I think there might be an issue then I look at the s/f values, which shows the time fast/slow per furlong for the race. As you see above they are all reasonably similar with no obvious split in the range covered - Hurdles are 1.23, 1.53. 1.64, 1.45, 1.75 while the chases are 1.25 and 1.38, which fit happily into the hurdle range towards the lower end. You might see 1.2 -1.6 for hurdles and 2.3 - 2.5 for the chases (as an example), to go along with seeing both chases are showing bigger OR-rtg scores, that would be a pointer that the chases were running in slower ground.

Additional help comes from the 'This runner' column (ratings the winner has previously earned) that allows me to see if the rating for today is a reasonable fit with previous ratings (ie not 50 lbs better than before, or 50 lbs worse) and the 'spd rtgs' column that shows me ratings of winners of similar races at the track, which helps show if the rating is unusual for this class/type/track. These two columns can help confirm an opinion formed by looking at the OR-rtg and s/f, I wouldn't base a decision on this last pair if I hadn't already spotted a potential issue from the s/f and OR-rtg,.

If I decide there's a problem I then split the results data I've already processed from the bare results file and treat chases and hurdles as two separate meetings, and see what comes out, checking to see if it likes a better fit.

There is a large degree of personal opinion involved in this, it's deciding if 'it looks right' - I might reprocess things several times, thinking each rerun still doesn't feel right, until I hit on something I can live with.

Dave

I'm currently trying to rate National Hunt using quadratic equations , done a few courses and my sheet calculates everything automatically.
Like Dave there is a column for each race that indicates a going allowance for that individual race (Horse GA) then it calculates a going allowance and ratings for the whole card and a going allowance and rating when the Hurdles and Chases have been split.
First looking at this meeting from Limerick 28/12/2018

racetime	track	yards	Type	horse_name	STANDARD	Comp	Horse GA	GA	Rating	SplitGA	SplitRating
13:40:00​	Limerick	3520​	Hurdle	Espoir Dallen (FR)	218.47​	22.53​	164​	36​	38​	166​	168​
12:30:00​	Limerick	3900​	Chase	Redhotfillypeppers (IRE)	260.83​	7.17​	31​	36​	134​	32​	130​
14:10:00​	Limerick	3520​	Hurdle	Press Conference (IRE)	218.47​	32.13​	181​	36​	0​	166​	104​
13:05:00​	Limerick	4960​	Chase	Lilshane (IRE)	331.60​	16.40​	58​	36​	104​	32​	100​
14:45:00​	Limerick	4620​	Hurdle	Owl Creek Bridge (IRE)	297.85​	44.65​	185​	36​	0​	166​	92​
12:00:00​	Limerick	4840​	Hurdle	Melly And Me (IRE)	314.37​	51.03​	193​	36​	0​	166​	79​
15:50:00​	Limerick	3520​	H Flat	Cobblers Way (IRE)	214.10​	37.90​	219​	36​	0​	166​	75​
15:20:00​	Limerick	4620​	Hurdle	Chosen Hour (IRE)	297.85​	50.85​	213​	36​	0​	166​	59​

The calculations are indicating if I take the meeting as a whole the going allowance is 36 (0.36s/f slow) but split it suggest the Chase was 0.32 s/f slow and the hurdles was much heavier at 1.66 s/f slow ( of course this could be because my standards are not lining up well , i'm still learning if they can be done well with quadratic equations).
This seems to be a no brainer if going on my data a clear split and if I did split Espoir Dallen (FR) would be rated 168( the highest hurdle rating I've seen so far in my sheets) and went on to win Champion Hurdle so that looks promising.
I looked at other Limerick meetings and the chase course is very often slower or on par with hurdles so i'm not concerned at this stage that it has returned lenient standards for the chase course.

Then there are the meetings where the chase and hurdle are very close to the point where it doesn't matter if you split them or not because they have roughly the same going allowances

Where i'm finding difficulty is meeting like Haydock 25/11/2017

racetime	track	yards	Type	horse_name	STANDARD	Comp	Horse GA	GA	Rating	Split GA	SplitRating
15:00:00​	Haydock	5643​	Chase	Bristol De Mai (FR)	398.83​	21.47​	120​	104​	162​	120​	179​
14:25:00​	Haydock	5017​	Hurdle	Sam Spinner	330.05​	17.55​	109​	104​	157​	103​	156​
12:40:00​	Haydock	6275​	Chase	Hainan (FR)	449.15​	35.85​	127​	104​	124​	120​	141​
13:50:00​	Haydock	4151​	Hurdle	Limited Reserve (IRE)	270.06​	16.94​	103​	104​	137​	103​	136​
13:15:00​	Haydock	4756​	Chase	Clan Des Obeaux (FR)	330.36​	28.24​	148​	104​	117​	120​	133​
15:35:00​	Haydock	5643​	Chase	Chase The Spud	398.83​	35.57​	143​	104​	107​	120​	124​
12:10:00​	Haydock	3444​	Hurdle	Delire Destruval (FR)	220.42​	20.58​	126​	104​	107​	103​	107​

The problem is the sheet is saying either 1.04 s/f slow for the meeting as a whole but if split 1.03 s/f slow for hurdles and 1.20 s/f slow for Chases, my first instinct here I think would have been to split and this means Bristol De Mai (FR) gets a rating of 179 (the biggest I've seen by about 3lbs on the sheets I've done so far) . He has just won a Grade 1 by 57 lengths so is this believable, now with hindsight I would say the going was the same give a going allowance of 1.04 s/f slow and rated Bristol De Mai (FR) 162 which I believe now is the correct answer, but I know I would have split at the time.
So trick to know what is the correct course of action and I imagine even harder to automate on a spreadsheet.
davejb TheBluesBrother would you have split the Haydock meeting at the time if you had the same data as above?

 

[TheBluesBrother]

davejb

DE just informed me that the National Hunt Chase at the Cheltenham Festival will be run on the old course over a new distance of 3m5f201y



Mike.

 

[TheBluesBrother]

First looking at this meeting from Limerick 28/12/2018

A word of warning about Limerick, when they hold their festival the rails are moved each day but the official race distances stay the same.

Note: from 2018 I now use my updated (-28 method) as a constant in calculating going allowances.

Where i'm finding difficulty is meeting like Haydock 25/11/2017

Mike.

 

[Outlander]

A word of warning about Limerick, when they hold their festival the rails are moved each day but the official race distances stay the same.

Note: from 2018 I now use my updated (-28 method) as a constant in calculating going allowances.

View attachment 81088



View attachment 81089

Mike.

So you split both at the time (bit easier decision at Haydock because your going allowances are further apart)
you rated both horses very highly for your ratings, you must have thought Espoir Dallen has a fair chance in the champion hurdle after that.
Thanks much appreciated

 

[davejb]

O Outlander
The other thing you need to watch for is when the going changes as a meeting progresses - more often than not this would be where you see s/f values increasing as the day goes by due to the going getting softer, but it can of course speed up a bit with sun and wind drying a surface. It's all 'best guess' to a degree, some days more so than others, but it helps sometimes to try to figure out what was happening in descriptive rather than numerical terms... ie 'it was raining heavily, the ground started soft but got cut up and the going allowance ought to be allowing for the slower ground as the meeting went by'. Once you have a theory of what you are looking at then it can be easier to work out how the numbers ought to be trending.

Dave

 

[TheBluesBrother]

When I work on northern NH meetings the one constant I always keep picking up on is how bad the quality of racing is, If the north has one winner at this year's Cheltenham festival I will be very surprised.

BENIE DES DIEUX is an exceptional mare whose fall at the final flight at Cheltenham last season remains the sole blot on her record since joining Willie Mullins. This was a formality once her main market rival capitulated, but nonetheless she impressed with the manner in which she grabbed control of the race. She would give Mullins an outstanding chance of a tenth Mares' Hurdle success, but may well have done enough here to merit a tilt at the Stayers' Hurdle. Mullins talked of her in glowing terms.

QUOTES: She could be better than any mare I've ever trained. You would seriously have to look at the Stayers' Hurdle after that performance - Willie Mullins, trainer of BENIE DES DIEUX.

TAVUS made no impact in the Grade 1 Finale Hurdle over Christmas, but his Newcastle form has received a big boost from the runner-up Navajo Pass. He won this quite comfortably under a penalty, albeit helped by the market leader's early departure, while his jumping wasn't especially slick.
Yesterday top NH speed figures:


HAREEM QUEEN, back on Fibresand, travelled comfortably just off the pace and saw her race out well. She's progressing nicely on the AW and may yet have more to offer.

HART STOPPER, wearing first-time cheekpieces, ran well to finish second in a good time in a similar Kempton contest 15 days earlier. He had to wait for a gap to emerge up the near rail and narrowly took his chance to strike with aplomb under Hayley Turner in the final furlong. It was his first AW victory at the eighth attempt.

Yesterday's Top AW speed figures:




My tracker today...


Mike.

 

[mlmrob]

When I work on northern NH meetings the one constant I always keep picking up on is how bad the quality of racing is,

I hope it stays that way as well as the amount of horses that are easy to dismiss makes life much easier.

 

[tacker]

When i think back to northern trainers such as STEVENSON, OLIVER, EASTERBY and DICKINSON it strikes me as being a very sad state of affairs but perhaps symptomatic of the uk.

 

[lazar23]

My tracker today...

Very bad, Mike...

 

[TheBluesBrother]

The hype machine was off and running after Glynn's 11 length romp in a class 5 novices hurdle yesterday and was put up as a fancy for the Supreme Novices where he is currently 25/1 or the Albert Barlett 20/1.
My immediate thoughts were WTF, he wins a Class 5 novices hurdle and is expected to step up into Grade 1 company, yesterday's speed figure of 91 was respectable on the day, the only problem I have is, does he have the ability to run to a figure of 130+ needed for Grade1 races at Cheltenham?

GLYNN beat a couple of subsequent British bumper scorers when winning his only Irish point in March. There wasn't much market confidence behind him on this rules debut, but he travelled smoothly, jumped with fluency, and cleared away for an impressive success. A stronger pace will see him to even better effect and he's a useful prospect.

QUOTES: I hope that looked as good as it felt. GLYNN gave me an incredible feel. He settled very well, jumped and travelled - I wouldn't be scared to drop him down to two miles. We were expecting a good run and he delivered with flying colours. He isn't the biggest in the world but he's very well put together - Jerry McGrath, jockey.

Yesterday's top NH speed figures:


Yesterday's top AW figures:




Mike.

 

[Paceman]

TheBluesBrother I ended up with a constant of 76lbs a slight difference to Mick Fitzgerald's 45lb.

So a horse capable of running to a figure of 100 on the Flat ought to be capable of running 176 over jumps? That seems rather high?

 

[Outlander]

TheBluesBrother I ended up with a constant of 76lbs a slight difference to Mick Fitzgerald's 45lb.

So a horse capable of running to a figure of 100 on the Flat ought to be capable of running 176 over jumps? That seems rather high?

Don’t think there is a correct answer to this question
I reckon flat to hurdles +35
Hurdles to flat - 50
Chase to flat - 65
Chase to hurdles -10
NHF to flat -35
But even that would make Kauto Star a group 1 horse on the flat, so it’s all just a rough guide really

 

[TheBluesBrother]

P Paceman

My calculation is one way only, jumps to the flat, based on a horses jumping ability.

The best "time" man in the country James Willougby, touched on this very subject in the Thoroughbred Daily New in 2015 where he stated that the flat scale was incorrect and needs to be increased by 25% to bring it into line with the national hunt, due to the fact that the NH handicapping is based on a wider scale.

As James Willoughby is a better mathematician than I am, this is good enough for me.

James Willoughby: Peter Willett Tribute/On The Handicap Scale

By James Willoughby It is my intention to dedicate this column to the memory of Peter Willett who died this week aged 96. I will never forget being invited to tea at his house in Rotherwick, Hampshire, following a spot-on critique he penned of one of my columns on handicapping in the Racing Post in

www.thoroughbreddailynews.com

Mike.

 

[TheBluesBrother]

It's good to see that Kempton has started using the inner course again (5f & 1m2f) especially when I recently sent an email to the clerk of the course BC asking him why the inner course was riding twice as fast as the outer, he never replied, he did stop using the inner course until yesterday.
Again the official going description yesterday was given out as standard to slow, not bloody once have I ever found that, the only true standard to slow going allowance I have ever seen was at Wolverhampton in Dec 2010 when they raced on the polytrack (course was frozen) and I had the going allowance at -0.25s/f, yesterday I had the going allowance at +0.42s/f (standard to fast).



*************************************************************

SANTINI the winner of the Cotswold chase earned a speed figure yesterday of 104 (best 113) he is being put up as a fancy for the Gold Cup, last years winner Al Bourn Photo earned a speed figure of 142 when winning last year, make your own mind up on his chances.

MISTER FISHER made it 2-3 over fences and, while enhancing his leading trainer's fine record in the race, confirmed himself as a big player for the Cheltenham festival. There was a lot to like about his response when asked to make a move from two out and he landing running after a fine leap at the last. He proved himself over 2m4f at Prestbury Park last time out, but obviously still has plenty of toe and odds of around 16-1 in a fairly open-looking Arkle look more than fair. With the possibility of more to come back up in trip, though, it was little surprise that he got promoted to the top of the market for the Marsh Chase. Betfred were most impressed and went 6-1 ante-post favourite.

QUOTES: MISTER FISHER is twice the horse of last season, he's very good over fences and he did everything perfect today. The ground is the key, it's a bit quicker today. It's been soft or dead all season and he's looked slow, on quicker he shows his foot - James Bowen, rider.

Yesterday's top NH speed figures:


CORAZON ESPINADO's stable won this last year and, under a strong drive from Luke Morris, he got his head in front for the first time at this course (second three times). History shows he doesn't want to go up much but he'll remain competitive here after a small rise.

POWER LINK's second last time came in a stronger contest than this, albeit at a furlong shorter, but he just couldn't put it to bed today. it was yet another good piece of form but 7f looks to be his trip as he travels so well, for all that he went very close.

SPIRIT POWER, a C&D winner who scored at Newcastle last time, was up 4lb but still just 1lb higher than when notching a career best at Southwell last February. He broke smartly from an awkward draw, bagged the rail and set his own fractions to score cosily. A bold hat-trick bid is anticipated.

Yesterday's top AW speed figures:




Speed figures:

AW: AW2018_2019.xlsx

NH: NH2018_2019.xlsx

Mike.

 

[tacker]

Henderson has just said mister fisher is a two and half mile horse and the reason he ran over two miles yesterday was simply the better ground.

 

[TC2]

TheBluesBrother what do you allow for rail movements. In terms of seconds and lengths?

 

[TheBluesBrother]

@TheBluesBrother what do you allow for rail movements. In terms of seconds and lengths?

For the national hunt I divide the yardage by 15 and when the flat turf starts I will use 13.

So if the rail movement increased the distance you subtract the time adjustment from the comparison per furlong time and likewise if the distance is short you add the time adjustment.

Sedgefield today:
Soft (Good To Soft In Places)
Going Stick 6.5 on Sun 6:40am
Rails: Bends 1,2 + 4 are divided, bend 3 is shared. All rails have been moved to provide fresh ground.

• 1:00pm: Race distance is now -16y to 2m 162y (+1.06s)
• 1:30pm: Race distance is now +18y to 2m 3f 83y (-1.2s)
• 2:00pm: Race distance is now -20y to 2m 3f 168y (+1.33s)
• 2:30pm: Race distance is now +18y to 2m 95y (-1.2s)
• 3:00pm: Race distance is now -24y to 2m 5f 10y (+1.6s)
• 3:30pm: Race distance is now +27y to 3m 2f 86y (-1.8s)
• 4:00pm: Race distance is now -16y to 2m 162y (+1.06s)

Mike.

 

[TheBluesBrother]

STORMY IRELAND, second in the Grade 1 mares' events at Cheltenham and Punchestown last season, was suited by the strong pace supplied by the eventual third and was far too good for these rivals. She asserted effortlessly from two out, and was unextended in making it three in a row for the season. She should be ready to put up another bold showing in March.

QUOTES: She's improving all the time and it was good to see her win so well over a trip probably shorter than ideal. She also showed that she doesn't have to make the running. I'd imagine she'll go for the Grade 1 mares at Cheltenham -Willie Mullins, trainer of STORMY IRELAND.

Today's Top NH speed figures:




Mike.

 

[TC2]

thankyou TheBluesBrother Thats similar to what i had.

Really interesting read. are your going allowances, what you've compiled over time?

 

[Steeplechasing]

Many thanks, Mike. I was very impressed visually by Mister Fisher. Good to see it backed up on time. He won't get good ground for the Arkle but would still have a chance if there is a fair bit more good than soft.

Protektorat ran a remakable race having been so keen for so long - still pulling two put. He'll top my list for whatever Cheltenham handicap he aims for.