The Power Of Ratings

A new look at stallion statistics as an aid to decision-making

Lightbulb760506I had a chance encounter the other day with a media mogul. Successful, wealthy, intelligent. Well, maybe the latter part of that statement may need examination, as he expressed a desire to get involved in racing and breeding. Of all things. Anyway, we got talking and he clearly knew a thing or two. Stallions, he said, have to be the key to success. And could I help him with solid background information in that respect. Give me something out of the ordinary, he added.

Interesting request. But where to start?

My area of expertise is rating horses on ability. I’ve done it for a good few decades, and I know the method works for racehorses. So if it works for them, why not for stallions? Take a large enough sample from a database, and see what happens when you take all progeny and rate them as a group. Would that reveal anything useful?

First a little background

In short, a rating is an assessment of the best ability of a horse, as shown on the track. Ability is inborn – you have what you’re given by your parents, no more, no less. Maturity, good training, sensible feeding – those are factors that bring out a horse’s inherent ability, but they won’t be able to improve on it. You either have it, or you don’t.

Most horses’ ability is average, for that’s how nature works: regression to the mean. Some horses may be better, some will be worse. But on average, the ability in a horse population remains quite constant from year to year. And a rating shows how much better or worse a horse is compared to the mean rating of the population as a whole.

For convenience, ratings are expressed in weight – pounds in the UK, and comparable half-kilo’s in most of the rest of the world.

Ratings can be applied in a wide variety of ways, foremost to frame handicap weights on a day-to-day basis. Creatively applied, ratings can inform about many different things and give them bite. Including, hopefully, stallion statistics.

The first thing to do is to establish benchmarks. Individual ratings would give too much variety in graphs, so the thing to do is make ratings ‘bins’, groups of ratings, in a statistically sensible way.

For this all stallions with runners in the 2013/14 season were selected, and for all of them all their runners from the last ten years, with their career-best ratings. The bins look like this, with the bar-graph showing the percentage of runners for all stallions in each bin.

Graph depicting all offspring of stallions, by ratings

All Offspring of Stallions, by ratings

All offspring of stallions by ratings – as a table

Rating bins <55 55-69 70-79 80-89 90-99 >99
All 19% 24% 27% 17% 9% 3%
Colts 18% 22% 26% 19% 10% 5%
Fillies 20% 27% 29% 16% 7% 2%

To give the rating bins perspective, all black type winners from the ten-year period were selected, with the ratings recorded with the wins (in other words, not the career-best ratings of the winning horses). The bar-graphs again reflect percentages in each bin.

Graph depicting Ratings of Gr1 winners

Ratings of Gr1 winners

Graph depicting Ratings of Gr2 winners

Ratings of Gr2 winners

Graph depicting Ratings of Gr3 winners

Ratings of Gr3 winners

Graph depicting Ratings of Listed winners

Ratings of Listed winners

The bar-graphs give a clear idea of what kind of rating is needed to win at the four levels of Graded and Listed races.
It appears that the 90-99 bin is a minimum requirement for bold caps black-type.
It also appears that the conventional wisdom that ‘Gr1 horses win Gr1 races’ in the main rings true (excluding 2yo’s).

.

Now for the real stuff

A selection of notable stallions is assembled below, to compare. The bar-graphs show the percentages of rated offspring in each bin for nine sires.

There are ten bars in all in each bin. On the left the population average (shown earlier on its own, in the first bar-graph), then next to it in alphabetical order nine stallions (their names and bar-colour next to the graph).

The stallions in the example are regulars in the top-sire stats. You’d expect their bars to be below the population-average (the blue bar on the left in each bin) in the lefthand side bins, and above it on the righthand side. Not unexpectedly, there are major differences between the nine stallions.

Graph depicting Offspring of Selected Stallions, by Ratings

Offspring of Selected Stallions, by Ratings

The bar-graphs give a good visual idea, but to combine bins it is easier to work with a table.

Offspring of selected stallions, by percentage in rating-groups – as a table

Rating bins <55 55-69 70-79 80-89 90-99 >99
All stallions 19% 24% 27% 17% 9% 3%
Captain Al 10% 16% 26% 27% 15% 6%
Dynasty 17% 21% 28% 17% 11% 5%
Fort Wood 9% 17% 29% 23% 16% 6%
Jet Master 10% 18% 25% 26% 14% 7%
Kahal 18% 25% 28% 17% 9% 4%
Silvano 14% 21% 28% 19% 10% 8%
Trippi 8% 21% 43% 14% 7% 7%
Var 9% 15% 32% 26% 11% 7%
Western Winter 7% 15% 27% 25% 18% 8%

Let’s do some maths and make observations.

Kahal is virtually the same as the population-average in all bins.

Silvano and Western Winter jointly lead in the top-bin, with double the population average. The pack is close on their heels, though.
When the two top-bins are combined, Western Winter is in a class of his own with 27%, followed by Fort Wood, Captain Al and Jet Master.
Combine the top three bins and you’ll find Western Winter on 52%, Captain Al 48%, Jet Master on 47%, Fort Wood 45% and Var 45%. The overall population-average for the three bins is 30%.

Then the reverse side of the coin, the low bins. The under-55 bin horse is unlikely to ever win a race, while the 55-69 bin is filled with mostly modest winners – if at all.
It’s again Western Winter who comes off best, with the lowest percentage in the lowest-rating bin. When the two lowest-rating bins are combined, the population average is 44%, this vs. 21% for Western Winter, who leads here, too.

When statistical methods are applied to a ratings population, there is a significant difference between the sexes in terms of mean. The difference equates to about 3 points – which is reflected in races as sex-allowance.
Rating-bins show dramatic changes when split by sex, also reflected in the population averages against which the sires must be measured.

Graph depicting Male Offspring of Selected Stallions, by Ratings

Male Offspring of Selected Stallions, by Ratings

Male offspring of selected stallions, by ratings – as a table

Rating bins <55 55-69 70-79 80-89 90-99 >99
All colts 18% 22% 26% 19% 10% 5%
Captain Al 9% 15% 23% 27% 16% 9%
Dynasty 13% 19% 27% 17% 16% 9%
Fort Wood 10% 16% 25% 24% 17% 8%
Jet Master 10% 14% 27% 26% 14% 9%
Kahal 18% 24% 23% 19% 10% 5%
Silvano 13% 18% 24% 18% 13% 14%
Trippi 11% 20% 39% 14% 8% 8%
Var 10% 13% 29% 29% 12% 7%
Western Winter 7% 11% 22% 23% 24% 12%

.

Graph depicting Female Offspring of Selected Stallions, by Ratings

Female Offspring of Selected Stallions, by Ratings

Female offspring of selected stallions, by ratings – as a table

Rating bins <55 55-69 70-79 80-89 90-99 >99
All fillies 20% 27% 29% 16% 7% 2%
Captain Al 10% 17% 29% 27% 13% 3%
Dynasty 22% 23% 29% 18% 6% 1%
Fort Wood 8% 18% 34% 22% 14% 3%
Jet Master 11% 21% 23% 26% 14% 6%
Kahal 18% 26% 32% 14% 7% 3%
Silvano 15% 24% 32% 19% 8% 3%
Trippi 6% 22% 47% 13% 6% 7%
Var 8% 17% 36% 23% 11% 6%
Western Winter 7% 19% 32% 27% 13% 3%

Sex Matters

The population percentage in the top-bin was 4% with sexes combined, but when split by sex the colts are 5% and fillies 2%. Combine the top-two bins, and the overall 13% (both sexes together) becomes 15% for colts, and 9% for fillies. Similarly, the top-three bins’ overall percentage of 30% becomes 34% for colts and 25% for fillies.

The conclusion is that in general terms fillies aren’t ‘as good’ racehorses as colts (could it be because most were gelded?).

Leaders in the top-bin for colts are Silvano and Western Winter, both more than twice as high as the overall population percentage for colts.

With fillies it is a different story. There the top-bin honours are disputed by Trippi, Jet Master and Var – all achieving three times the population average for their sex. Things change further when the top-two bins are combined.

For the colts Western Winter is a clear leader, the only sire to more than double the overall average for the sex. Some way behind him follow Captain Al, Dynasty, Silvano and Fort Wood. With three bins Western Winter stays well ahead of Captain Al, and the closely grouped Fort Wood, Jet Master and Var.

In the fillies’ top-two bins Jet Master doubles the female population average, followed by Fort Wood, Var, Captain Al, and Western Winter. With three bins the picture remains the same.
.

Fun With Numbers

So is there a real use with this for our inquisitive media mogul?

Presumably, if he had a shortlist of comparable yearlings to buy, this information can help in making the final decision. Or if he wanted to buy a broodmare, or breed a mare to a shortlist of stallions with the objective of hitting a big one, then for that too the information could help.

The real point is that rating evaluation provides insight which cannot be gleaned from commonly used stats, such as those based on seasonal earnings or black type performance.

If you’d like to see it all, here’s a list of sires who had runners last season, with performance of offspring going back a decade. To add another level, they have been grouped by year of first crop, and then alphabetically within those groups. Your thoughts, discoveries, and comments are appreciated.

– Karel Miedema

Offspring of sires by rating

Stallions <55 55-69 70-79 80-89 90-99 >99
All stallions 19% 24% 27% 17% 9% 3%
First crop 1994
Fort Wood 9% 17% 29% 23% 16% 6%
First crop 1998
Western Winter 7% 15% 27% 25% 18% 8%
First crop 1999
Muhtafal 14% 21% 26% 24% 11% 4%
First crop 2002
Captain Al 10% 16% 26% 27% 15% 6%
Century Stand 23% 35% 26% 13% 1% 2%
Count Dubois 21% 20% 30% 19% 8% 3%
Jet Master 10% 18% 25% 26% 14% 7%
Kahal 18% 25% 28% 17% 9% 4%
Lecture 17% 29% 30% 14% 7% 3%
First crop 2003
Almushtarak 35% 22% 27% 14% 2% 0%
Announce 14% 32% 30% 14% 9% 1%
Asaal 33% 22% 22% 13% 9% 0%
Doowaley 18% 27% 26% 18% 7% 3%
Indigo Magic 23% 28% 27% 18% 4% 0%
Mogok 20% 23% 30% 17% 7% 2%
The Sheik 28% 24% 28% 13% 6% 1%
Windrush 20% 25% 28% 18% 6% 2%
First crop 2004
Badger’s Drift 15% 26% 35% 16% 5% 3%
Silvano 14% 21% 28% 19% 10% 8%
First crop 2005
Bezrin 26% 27% 26% 13% 4% 4%
Malhub 36% 27% 20% 9% 7% 1%
Sarge 20% 34% 30% 11% 4% 1%
First crop 2006
Alado 24% 41% 24% 0% 10% 0%
Dupont 28% 27% 22% 18% 3% 2%
Dynasty 17% 21% 28% 17% 11% 5%
Spectrum 21% 29% 29% 14% 6% 1%
Toreador 21% 32% 26% 14% 7% 1%
Var 9% 15% 32% 26% 11% 7%
First crop 2007
Battersea Park 38% 31% 23% 0% 8% 0%
Black Minnaloushe 25% 27% 26% 14% 7% 2%
Damage is Done 60% 10% 20% 0% 10% 0%
Greys Inn 23% 28% 24% 16% 6% 2%
Jam Alley 43% 27% 20% 3% 7% 0%
King of Kings 30% 30% 25% 11% 3% 1%
Lundy’s Liability 20% 40% 18% 12% 7% 2%
Miesque’s Approval 23% 31% 30% 12% 5% 0%
Overlord 21% 33% 33% 9% 4% 0%
Right Approach 20% 29% 29% 13% 7% 2%
Scripture 75% 25% 0% 0% 0% 0%
Solskjaer 26% 39% 18% 12% 3% 2%
Stagelight 28% 44% 17% 11% 0% 0%
Tiger Dance 24% 32% 31% 7% 4% 1%
Way West 35% 31% 23% 5% 5% 0%
Whitechapel 42% 33% 17% 6% 3% 0%
First crop 2008
Go Deputy 21% 31% 24% 15% 6% 3%
Tobe Or Nottobe 37% 28% 28% 7% 0% 0%
First crop 2009
Ashaawes 29% 24% 26% 11% 8% 1%
Atso 36% 45% 11% 7% 1% 0%
Imperial Stride 32% 27% 22% 13% 7% 0%
Lion Tamer 35% 29% 21% 15% 0% 0%
Mullins Bay 20% 34% 30% 13% 2% 1%
Ravishing 41% 33% 16% 8% 2% 0%
Soar With Eagles 32% 35% 18% 12% 3% 0%
St Petersburg 43% 35% 14% 6% 2% 0%
Trippi 8% 21% 43% 14% 7% 7%
First crop 2010
A P Arrow 32% 48% 16% 2% 2% 0%
Admire Main 26% 29% 39% 3% 0% 3%
Antonius Pius 24% 26% 27% 16% 8% 0%
Argonaut 35% 30% 25% 6% 3% 0%
Daring Bid 59% 29% 6% 6% 0% 0%
Horse Chestnut 13% 27% 34% 17% 8% 2%
Jay Peg 19% 37% 34% 8% 3% 0%
Judpot 13% 30% 39% 11% 5% 2%
Kildonan 11% 37% 22% 19% 11% 0%
Rebel King 29% 30% 25% 16% 0% 0%
Royal Air Force 23% 46% 15% 15% 0% 0%
Sail From Seattle 22% 26% 33% 12% 5% 2%
Seul Amour 46% 15% 31% 0% 8% 0%
Spanish Harlem 30% 45% 25% 0% 0% 0%

 

Male offspring of sires by rating

Stallions <55 55-69 69-79 79-89 89-99 >99
All COLTS 18% 22% 26% 19% 10% 5%
First crop 1994
Fort Wood 10% 16% 25% 24% 17% 8%
First crop 1998
Western Winter 7% 11% 22% 23% 24% 12%
First crop 1999
Muhtafal 13% 17% 28% 25% 11% 7%
First crop 2002
Captain Al 9% 15% 23% 27% 16% 9%
Century Stand 23% 33% 25% 13% 2% 4%
Count Dubois 16% 19% 31% 24% 7% 3%
Jet Master 10% 14% 27% 26% 14% 9%
Kahal 18% 24% 23% 19% 10% 5%
Lecture 16% 24% 28% 21% 7% 5%
First crop 2003
Almushtarak 37% 17% 24% 17% 5% 0%
Announce 14% 29% 27% 19% 9% 1%
Asaal 31% 19% 26% 10% 14% 0%
Doowaley 17% 21% 24% 24% 9% 4%
Indigo Magic 22% 31% 25% 18% 5% 0%
Mogok 20% 19% 29% 18% 10% 4%
The Sheik 30% 21% 29% 15% 5% 1%
Windrush 21% 22% 27% 21% 6% 3%
First crop 2004
Badger’s Drift 12% 31% 29% 16% 6% 6%
Silvano 13% 18% 24% 18% 13% 14%
First crop 2005
Bezrin 19% 25% 27% 15% 5% 9%
Malhub 31% 23% 22% 12% 9% 2%
Sarge 29% 26% 26% 11% 9% 0%
First crop 2006
Alado 7% 47% 33% 0% 13% 0%
Dupont 18% 25% 24% 25% 3% 4%
Dynasty 13% 19% 27% 17% 16% 9%
Spectrum 22% 25% 27% 16% 9% 2%
Toreador 17% 33% 26% 15% 8% 2%
Var 10% 13% 29% 29% 12% 7%
First crop 2007
Battersea Park 44% 33% 11% 0% 11% 0%
Black Minnaloushe 18% 26% 27% 18% 8% 4%
Damage is Done 100% 0% 0% 0% 0% 0%
Greys Inn 19% 28% 23% 14% 12% 4%
Jam Alley 42% 26% 21% 5% 5% 0%
King of Kings 30% 26% 30% 9% 3% 2%
Lundy’s Liability 23% 31% 15% 15% 10% 5%
Miesque’s Approval 21% 30% 29% 13% 6% 1%
Overlord 18% 31% 36% 10% 5% 0%
Right Approach 15% 26% 25% 17% 13% 4%
Scripture 75% 25% 0% 0% 0% 0%
Solskjaer 31% 36% 20% 7% 2% 4%
Stagelight 14% 57% 29% 0% 0% 0%
Tiger Dance 17% 34% 37% 2% 7% 2%
Way West 42% 26% 26% 3% 3% 0%
Whitechapel 25% 37% 19% 12% 6% 0%
First crop 2008
Go Deputy 14% 30% 25% 18% 9% 5%
Tobe Or Nottobe 33% 31% 26% 10% 0% 0%
First crop 2009
Ashaawes 26% 26% 26% 9% 14% 0%
Atso 34% 43% 11% 9% 3% 0%
Imperial Stride 27% 27% 13% 20% 13% 0%
Lion Tamer 29% 41% 6% 24% 0% 0%
Mullins Bay 22% 37% 30% 6% 3% 2%
Ravishing 41% 30% 15% 11% 4% 0%
Soar With Eagles 36% 32% 18% 9% 5% 0%
St Petersburg 44% 32% 12% 8% 4% 0%
Trippi 11% 20% 39% 14% 8% 8%
First crop 2010
A P Arrow 32% 48% 16% 4% 0% 0%
Admire Main 25% 17% 50% 8% 0% 0%
Antonius Pius 26% 21% 17% 26% 11% 0%
Argonaut 29% 32% 24% 11% 5% 0%
Daring Bid 55% 27% 9% 9% 0% 0%
Horse Chestnut 18% 36% 24% 12% 6% 3%
Jay Peg 12% 37% 37% 12% 2% 0%
Judpot 6% 31% 37% 11% 11% 3%
Kildonan 7% 27% 33% 20% 13% 0%
Rebel King 36% 22% 25% 17% 0% 0%
Royal Air Force 40% 40% 20% 0% 0% 0%
Sail From Seattle 13% 29% 19% 19% 13% 6%
Seul Amour 50% 17% 17% 0% 17% 0%
Spanish Harlem 30% 30% 40% 0% 0% 0%

 

Female offspring of sires by rating

Stallions <55 55-69 69-79 79-89 89-99 >99
All FILLIES 20% 27% 29% 16% 7% 2%
First crop 1994
Fort Wood 8% 18% 34% 22% 14% 3%
First crop 1998
Western Winter 7% 19% 32% 27% 13% 3%
First crop 1999
Muhtafal 16% 26% 25% 23% 10% 1%
First crop 2002
Captain Al 10% 17% 29% 27% 13% 3%
Century Stand 23% 37% 25% 15% 0% 0%
Count Dubois 26% 21% 29% 14% 8% 3%
Jet Master 11% 21% 23% 26% 14% 6%
Kahal 18% 26% 32% 14% 7% 3%
Lecture 18% 33% 32% 9% 7% 1%
First crop 2003
Almushtarak 33% 26% 30% 10% 0% 0%
Announce 15% 34% 34% 9% 9% 0%
Asaal 36% 28% 16% 20% 0% 0%
Doowaley 21% 35% 28% 12% 4% 1%
Indigo Magic 25% 25% 29% 17% 4% 0%
Mogok 21% 27% 30% 16% 4% 1%
The Sheik 24% 29% 27% 11% 8% 0%
Windrush 19% 27% 30% 16% 6% 2%
First crop 2004
Badger’s Drift 18% 21% 42% 16% 3% 0%
Silvano 15% 24% 32% 19% 8% 3%
First crop 2005
Bezrin 33% 28% 26% 11% 2% 0%
Malhub 41% 32% 18% 4% 5% 0%
Sarge 13% 40% 33% 11% 0% 2%
First crop 2006
Alado 43% 36% 14% 0% 7% 0%
Dupont 40% 29% 19% 9% 3% 0%
Dynasty 22% 23% 29% 18% 6% 1%
Spectrum 20% 34% 31% 12% 3% 0%
Toreador 25% 30% 25% 13% 6% 1%
Var 8% 17% 36% 23% 11% 6%
First crop 2007
Battersea Park 25% 25% 50% 0% 0% 0%
Black Minnaloushe 31% 28% 25% 10% 6% 0%
Damage is Done 43% 14% 29% 0% 14% 0%
Greys Inn 27% 29% 25% 19% 0% 0%
Jam Alley 45% 27% 18% 0% 9% 0%
King of Kings 30% 34% 21% 12% 2% 1%
Lundy’s Liability 18% 48% 20% 9% 5% 0%
Miesque’s Approval 26% 32% 31% 9% 3% 0%
Overlord 28% 39% 28% 6% 0% 0%
Right Approach 25% 32% 33% 9% 2% 0%
Scripture 75% 25% 0% 0% 0% 0%
Solskjaer 22% 41% 16% 16% 5% 0%
Stagelight 36% 36% 9% 18% 0% 0%
Tiger Dance 33% 30% 22% 15% 0% 0%
Way West 30% 34% 21% 7% 7% 0%
Whitechapel 55% 30% 15% 0% 0% 0%
First crop 2008
Go Deputy 29% 32% 24% 12% 3% 0%
Tobe Or Nottobe 40% 26% 30% 5% 0% 0%
First crop 2009
Ashaawes 32% 22% 27% 14% 3% 3%
Atso 38% 46% 10% 5% 0% 0%
Imperial Stride 37% 27% 30% 7% 0% 0%
Lion Tamer 41% 18% 35% 6% 0% 0%
Mullins Bay 16% 30% 30% 23% 0% 0%
Ravishing 41% 35% 18% 6% 0% 0%
Soar With Eagles 25% 42% 17% 17% 0% 0%
St Petersburg 42% 38% 15% 4% 0% 0%
Trippi 6% 22% 47% 13% 6% 7%
First crop 2010
A P Arrow 32% 48% 16% 0% 3% 0%
Admire Main 26% 37% 32% 0% 0% 5%
Antonius Pius 23% 30% 37% 5% 5% 0%
Argonaut 41% 29% 27% 2% 0% 0%
Daring Bid 67% 33% 0% 0% 0% 0%
Horse Chestnut 6% 16% 45% 23% 10% 0%
Jay Peg 28% 36% 31% 3% 3% 0%
Judpot 18% 29% 41% 10% 0% 2%
Kildonan 17% 50% 8% 17% 8% 0%
Rebel King 24% 35% 25% 16% 0% 0%
Royal Air Force 12% 50% 12% 25% 0% 0%
Sail From Seattle 26% 24% 42% 8% 0% 0%
Seul Amour 43% 14% 43% 0% 0% 0%
Spanish Harlem 30% 60% 10% 0% 0% 0%

 

Have Your Say - *Please Use Your Name & Surname

Comments Policy
The Sporting Post encourages readers to comment in the spirit of enlightening the topic being discussed, to add opinions or correct errors. All posts are accepted on the condition that the Sporting Post can at any time alter, correct or remove comments, either partially or entirely.

All posters are required to post under their actual name and surname – no anonymous posts or use of pseudonyms will be accepted. You can adjust your display name on your account page or to send corrections privately to the EditorThe Sporting Post will not publish comments submitted anonymously or under pseudonyms.

Please note that the views that are published are not necessarily those of the Sporting Post.

Subscribe
Notify of
guest
10 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments

Share:

Facebook
WhatsApp
Twitter

Popular Posts