PEDIGREES AND STATISTICS

[pippex]

In what position of 4 generation horse pedigree is the name of a key sire most important..

Dave Dink studied thousands of pedigrees to attempt to find an answer..

His conclusion : the name of a key sire is most effective at the bottom of the pedigree in the 3rd or 4th generation..that is as sire of the 2nd dam or the 3rd dam...

so if he is right you should be looking for Northern Dancer(or whatever sire you choose) in 1 of these two positions..

I wont give the full stats..but at the time of his survey only 7% of horses had one of the top 20 sires as sire of the second dam and these horses had significantly better results..

The full story is in chapter 9 of "Racehorse Breeding Theories" by Frank Mitchell

[jth]

At last something i can get my teeth into Pippex

Dave Dink you cannot be serious.

No i read that chapter Pippex--and you can see Mr Dink flirting with Girl Power --but fell short --peer pressure, accepted breeding theory, whatever he flunked the issue :

ive looked at more pedigrees than is good for me-and certainly more than Dinky--
and having tried to use them in the war zone rather than as an academic exercise / research proposal to help the 'breeding / fleece industry .

You wont get stats from me on this im afraid. But maybe youll be able to use my answer to design quantitative tests to help me/you to confirm that GP is real and out there picking winners.

Here we go:
Northern Dancer:
name me one greatfilly he sired?
he sired all these males El Gran senor, Nijinsky, Ajdal,sadlers wells, lomond, secreto, The Minstrel and shareef dancer all great and all MALE

Lets call the 2nd dam (-2) , 3rd dam (-3) etc

in my experience the closer the dam is to the horse in ? the better:Whatever theory of genetic transfer you favour the closer that genetic influence the better-But i think all the dam line should be looked at. In Punting practice i go to -6 or -7 Max
Where the sire comes in , at least for me , is as a marker only

If you have a great sire --and the product of his union with D(-4) is GOOD or BAD in racing terms (EPS_Group race performance whatever)--then if its bad that means that dam is likely to be bad if its good then the dam could be good. But if you have a dodgy sire crossed with the same dam (-4) and the union is successful --thats more indicative that that dam carries the genetic oomph.

So to me --Ill look at the offspring of any dam and Ill knock her down if she falls short when crossed with a good sire but I'll build her up more than i knocked her down if she can pull one or hopefully more out of dodgy sire.

I look at the sire line from a FS point of view and thats all--when i look at a horses pedigree i concentrate exclusively on the dam tail line.
GP says --stallions bring them on--they contribute : but Mum knows best

[pippex]

Thanks Jth...

Very thought provoking response...

I think we could have something here

I put up the post for discussion purposes....

If you have a great sire --and the product of his union with D(-4) is GOOD or BAD in racing terms (EPS_Group race performance whatever)--then if its bad that means that dam is likely to be bad if its good then the dam could be good. But if you have a dodgy sire crossed with the same dam (-4) and the union is successful --thats more indicative that that dam carries the genetic oomph.

we will have to have a measure for sires..the aei and ci are well known ones..

So to me --Ill look at the offspring of any dam and Ill knock her down if she falls short when crossed with a good sire but I'll build her up more than i knocked her down if she can pull one or hopefully more out of dodgy sire.


If I read you right we need a new index for each dam..something similiar to the COMPARABLE INDEX* for sires..only reversed..I have to think this out


What we can do for each horse is taken each dam in turn and project what type of performance we should get on average from mating with a particular sire..its the reverse of measuring a sire by the quality of mares mated with..
we will have to sum up for all offspring..a type of a/e if you like for each dam..we could sum the aei's or ci's for each sire the dam is mated with and obtain what level of race performance we should get..then compare the
actual preformance with the projected performance..

repeat for the 2nd dam..3rd dam etc...apply a weighting on how far back in pedigree..

Feel free to blast this down Jth

*Comparable Index (Ci)
Indicates the average earnings of progeny produced from mares bred to one sire when these same mares are bred to other sires. A CI of 1.00 is considered average, 2.00 is twice the average, 0.50 half the average, etc.

[pippex]

we should use the damsireaei's ..I have a spreadsheet of these from previous work..

a big snag is that damsireaei's are updated monthly by equineline..so if we are going back a while we wont know the damsire's aei at that particular time..

[jth]

_

What we can do for each horse is taken each dam in turn and project what type of performance we should get on average from mating with a particular sire..its the reverse of measuring a sire by the quality of mares mated with..
we will have to sum up for all offspring..a type of a/e if you like for each dam..we could sum the aei's or ci's for each sire the dam is mated with and obtain what level of race performance we should get..then compare the
actual preformance with the projected performance..

repeat for the 2nd dam..3rd dam etc...apply a weighting on how far back in pedigree..

Feel free to blast this down Jth

Its a bit late to be thinking about this Pippex--the thing is there are (guess) 20 dams to every sire
If BR is true and i think it is you should eliminate all BR's >8
so stats ( a numbers game) probably has no application
You have to take a view on each horse as it presents itself-rely on PD information-look for tail line conduit mares/ high AEI mares and high AEI for the progeny and USE the as you say- the average (resultd) of mating with a particular sire ( the real McCoy ones) as a check

[jth]

we should use the damsireaei's ..I have a spreadsheet of these from previous work..

a big snag is that damsireaei's are updated monthly by equineline..so if we are going back a while we wont know the damsire's aei at that particular time..

I like Damsire AEI's--to me a Sire can best be judged by his damsire ability

i dont see that knowing the damsires AEI in the past or any particular moment in time is any particular help/ hindrance.

You seem to want to drag everything back to the sires-Damsire or straight sire (understandably from a stats viewpoint)

I want to get a reliable and predictive cofficient from the mums using the sires as a steadying reference

[pippex]

Just give me a bit of time to think about this..

I have to try a few approaches..

[pippex]

It will be possible to work out some sort of index for each dam,based on her racecourse performance and the racecourse performance of her offspring to date,using a pen and paper..this is very laborious though and I cant see any way at present of using computers to speed it up..

if all the data could be in the one place in the one sheet or database then we could make progress...

[0909dylan]

If I can bring this back to BREEDING THEORY FOR DUNCES. ........If a SIRE has had 20 races in its racing life......10 @ 7 Furlongs,WON them all.......10 races @ other distances,Lost them all................DAM ( or DAMSIRE,you choose )........Had 20 races in her/his racing career.........10 @ 7 furlong,WON them all......10 @ other distances.Lost them all.Whats the chance of the prodginy being 7 Furlong specialists ??????

[jth]

Pippex
Before thinking about working out a dam GP index .I think we need to be sure, or at least agree, that we are on the right track i.e that the damside is the golden goose and that within the dam tail line are the golden eggs.

The damsires of each of these dams --can provide a test of how good that dam is:

As i said earlier and now in attempt to be clearer by using an example --if we have Secretariat or Buckpasser (or some other accepted Supersire and damsire) on the dam tail line and we see the union of this Supersire with a dam produces only dud daughters then we can be pretty sure that the dam was and not a golden goose. if this occurs at D-4or D-3 (4th or 3rd generation) it is highly unlikely that the golden gooseishness (sorry .) can be recovered.

In the same way we would award extra GG points if a great dam comes out of a really Awful sire-im trying to think of an example.Say Chillibang.

conduit mares--present an interesting case-sometimes they seem to be good producers but not always --and its always nice to see a whole string of conduits on the dam tail line.

In short : Would be very interested in your further views and those of anyone else .

[jth]

If I can bring this back to BREEDING THEORY FOR DUNCES. ........If a SIRE has had 20 races in its racing life......10 @ 7 Furlongs,WON them all.......10 races @ other distances,Lost them all................DAM ( or DAMSIRE,you choose )........Had 20 races in her/his racing career.........10 @ 7 furlong,WON them all......10 @ other distances.Lost them all.Whats the chance of the prodginy being 7 Furlong specialists ??????

better than even money

But only if both parents were high class 7 f specialists and pretty likely they were high class if they won 10 races each.

If are you asking if race distance preference is hereditable--i think it is-so does flatstats and dosage theory.Just do a Flatstat table stat for each distance for sires OR damsires
If you are asking is 7f a special distance--most of the studies ive read say it is --up to 6F you can do without switching fully ino aerobic metabolism (if you're a horse that is )--7f marks the metabolic switch over

[0909dylan]

CHEERS.Not specifically 7 furlong......just asking generally.So if you made 3 lists of all SIRES + DAMSIRES + DAMS.......and say cropped the top 10% who performed best AS SPRINTERS DURING THEIR RACING CAREERS..........all the resultant offspring of matings from these lists you would expect to be GOOD SPRINTERS ???????...........7 furlong was a bad example......its supposed to be a "unique" distance in breedingspeak.............I believe a 7 furlong specialist is a rare thing..............e.g. PHLUKE

[jth]

no Dylan--i just look at what are good horses/ bloodlines more or less regardless of distance--although i dont want the real marathon runner gurneys given the choice.But i realise that doesnt answer your ?? Ill try again later

[pippex]

I agree tail line is place to look..cant be sure


The damsires of each of these dams --can provide a test of how good that dam is:

[jth]

nice, v nice

i cant see how you got those damsire AEIs from the equineline site pippex

i studied the 4 pedigrees you offered and all the others -the only others worth a further look are Sky gate and eastern warrior.

of your 4 Barwell Bridge wins a street by my analysis=all the tail dams down to -5 were good racers espstylish starand all produced good winners

Caravan of dreams had the best mum even though Smarten had a DS AEI of only 1.16

If Sir George can overcome his draw and if anyone can he can then that is the winner.

but Kirks record with resters is v poor and the price is rubbish as yet