Announcing the All New Key Ancestors Report

re: Announcing the All New Key Ancestors Report

Ian Tapp — Mon, 15 Jul 2013 14:42:48 GMT

Hi Mary,

Yes, absolutely. In the Key Ancestors Report with Analysis you are able to select only Australian stallions, or any mix of potential stallions that you want. You can search stallions by location or stud fee range, and then add/delete as you'd like.

re: Announcing the All New Key Ancestors Report

Mary — Sun, 14 Jul 2013 05:23:44 GMT

Can potential stallions be tailored to stallions standing in Australia only?

re: Announcing the All New Key Ancestors Report

Ian Tapp — Wed, 30 Jan 2013 21:58:56 GMT

Hi Al,

You can add stallions from both England and Ireland to your group of potential stallions. No need to run a second report. Just search first for stallions in England (with your desired stud fee range), and then do the same in Ireland. You can also add or delete individual stallions.

Email me at ian@truenicks.com if you have any trouble doing this, and I'll run the report for you.

re: Announcing the All New Key Ancestors Report

Al Ross (UK) — Wed, 30 Jan 2013 21:50:38 GMT

Hi Ian,

From the dropdown-box I note you can request the Key Ancestors Report by way of location which is an excellent facility when reviewing commercial considerations. For myself, I am interested in a report for stallions standing in England AND Ireland. Would this mean I have to pay for two separate reports or is there a facility for comparing stallions from two different regions in one report?

re: Announcing the All New Key Ancestors Report

Jay Ardee — Wed, 23 Jan 2013 20:25:24 GMT

How would these reports help me if I have a share in a stallion and am looking to buy or lease mares that would complement his ancestry?

In that case, I would be interested in a package rate that would allow me to experiment with multiple mares. Would you consider doing that? Or, since I assume that you can access these reports internally at a small cost, would you consider offering consulting on these types of questions?

re: Announcing the All New Key Ancestors Report

Tom Rolfe — Wed, 23 Jan 2013 14:44:57 GMT

Ian,

Based on your respose:

"The Key Ancestors Report considers the mare's entire pedigree (not just the broodmare sire line) to find even more specific data to help plan matings."

OMG! You're telling me that the TrueNicks rating is ONLY LOOKING AT THE BROODMARE SIRE LINE of my mare --- that's it? That means that all the genes are coming from my mare's sire-line? Although I'm not a genetisist, but wouldn't there be genes coming from my mare's dam --- like 50% of her genetic makeup? How could a system create an accurate rating without analysing 50% of my mare's pedigree? I have been relying on these nick ratings for years, and had no idea. I even fired my pedigree/bloodstock advisor, and have my in-house accountant do all my matings using TrueNicks!!!

How are you all going to educate all thoses breeders like me that have relied entirely on the TrueNicks Ratings?

Sadly,

Tom Rolfe

re: Announcing the All New Key Ancestors Report

Ian Tapp — Tue, 22 Jan 2013 22:43:39 GMT

Hi "Tom",

Your sarcasm is well taken—obviously there is a lot more to planning matings than simply checking a rating. The nick rating has been used for a long time in the industry and is very popular, but it's just one research tool to consider. The TrueNicks rating gives you the general success rate for the sire line/broodmare sire line combination. The Key Ancestors Report considers the mare's entire pedigree (not just the broodmare sire line) to find even more specific data to help plan matings. It can help differentiate between a group of stallions with equally high nick ratings, or it may find stallions whose nick ratings are only average but carry ancestors that are positive for your mare.

In addition to pedigree data, there are many other important considerations… physical/type, aptitude, commercial, etc. when planning a mating.

re: Announcing the All New Key Ancestors Report

Tom Rolfe — Mon, 21 Jan 2013 19:16:51 GMT

Hey Guys,

My mare is already a A+++ with a particular stallion --- it was free and easy. It took 30-seconds to plan my mating. If the TrueNick rating is based on genetics, correlated and statistically significant; why would I want to actually pay money and spend time on a Key Ancestors Report when I already know the stallion I have picked out for my mare is already AWESOME and off the charts? - Tom

re: Announcing the All New Key Ancestors Report

Jman — Thu, 17 Jan 2013 22:21:36 GMT

Greetings guys, this sounds like some of the things we chatted about last summer. To me, it offers a great second tool to the sire line nicks, with a lot of potential to drill down further. For example, I would be interested in seeing occasions of 'deal killers' in a pedigree, where perhaps a dam showing up in the pedigree creates a truly negative outcome, even when her daughters and granddaughters bred to the best stallions. I'm sure there are other things that could be unearthed, like your observation on Groom Dancer above. I don't know that an individual report for a single mare would necessarily allow such an analysis though.

re: Announcing the All New Key Ancestors Report

John from Baltimore — Wed, 16 Jan 2013 18:19:04 GMT

The fact ic your never going to convince me. However, let's assume there are people out there who are convinced and are willing to spend thier money to buy your report. What percentage of performace improvement can they expect to see from using this program and please explain in quantitative detail how you arrived at this number.

re: Announcing the All New Key Ancestors Report

Alan Porter — Wed, 16 Jan 2013 15:33:19 GMT

John from Baltimore,

I wouldn't make any claim to predict which genes are going to come forward, and from where. However, there are two factors that make things a lot less random than you might think.

Firstly, as Ian says, there is a very small percentage of the genome that contributes to racing performance. Secondly, in the thoroughbred horse blocks of genes are passed on together very consistently (look up "linkage disequilibrium ") which tends to mean that if you get A, you are also very likely to get B and C as well.

It's probably because of this that you do get ancestors, even those someway back in the pedigree, that can be shown to have a statistically significant effect. It is not so much of a question of genes necessarily being transmitted by this one ancestor (although they can be - look at grey coat color), but that when you have good descendants of an ancestor, other things in the pedigree tend to either reinforce or complement the gene variants that significantly impact racing performance. So the distant ancestor may not exert a direct influence, but more acts as a signpost.

It is not just the TrueNicks Key Ancestors report - which uniquely gathers a group of similarly-bred mares - but studies involving stallions (both by myself and others) that have found the same thing. That even if you take a horse with 1000+ runners, there will often be ancestors as far back as the fifth or sixth generation that independently (that is not through multiple channels) have a statistically significant positive or negative impact on a stallion (I would express this as being strongly associated with, rather than causal, but it's there all the same).

The TrueNicks rating and the Key Ancestors report may not be perfect - no rule based system can be - but looking at it from the personal standpoint of having studied pedigrees for about 40 years, and planned matings for close to 30 years, I can tell you that I would much rather have the data these programs provide than work without it.

re: Announcing the All New Key Ancestors Report

Ian Tapp — Wed, 16 Jan 2013 04:47:51 GMT

John from Baltimore,

For as many genes as there are in the genome, there's a relatively small subset of genes linked to racehorse performance. But when mapping your mare's genome and knowing the complete genetic and phenotypic profile of sire and dam is not possible, we use statistics to describe trends in matings.

While the Key Ancestors Report doesn't include a guarantee of success, it does provide real data of how closely related mares have produced so that you can develop mating strategies for your mare.

A pedigree isn't about any one particular ancestor; it's about the sum of the parts. It's also about probability, which you referenced in your last sentence about inheriting good or bad genes. The idea of this report is that you can maximize your probability of a good mating by maximizing the collective influence of ancestors present in successful matings of related mares.

Just as the horse with the highest Beyer Speed Figure doesn't always win the race, our report isn't a silver bullet, but is a research tool for breeders to educate themselves and make better decisions.

re: Announcing the All New Key Ancestors Report

John from Baltimore — Wed, 16 Jan 2013 03:02:39 GMT

Please enlighten on how many genes there are in the horse's genome and how the sequencing fom generation to generation is not so random that your program can pick winners and losers from the third generation and beyond. All one has to do is look at full brothers and sisters and see how different they can be because the genes don't fall in the same order. Yet, you want me to believe you can predict which genes good or bad a relative is going to recieve.

re: Announcing the All New Key Ancestors Report

Ian Tapp — Tue, 15 Jan 2013 23:50:46 GMT

Hi John from Baltimore,

Interesting analogy, but slightly too much hyperbole.

Breeders should absolutely consider the physical and aptitude of sire and dam in matings—overlooking such vital information would be irresponsible. The data in this report is not intended to replace essential horsemanship, but rather to supplement it.

You are correct that each successive generation loses more and more genetic impact; this is accounted for in the Key Ancestors Report. Two mares with only one 3rd-generation ancestor in common share just 1.56% relatedness. However, this report finds related mares that are 15-50% related to your mare (depending on your mare). This subset shares more genetic similarity than any other group of mares, and their combined produce records will describe any trends and affinities that the mares have established.

To steal your lottery analogy, we don't expect this report to guarantee you the Powerball jackpot, but it may reveal a few of the numbers.

re: Announcing the All New Key Ancestors Report

John from Baltimore — Tue, 15 Jan 2013 23:23:06 GMT

For a prospective foal one should only look at the parent's physical and grand parents and thier race records. After that the genes are so mixed up it becomes meaningless.

Since a horse is made up of a random series of genes that mix every mating, the further you go back the greater the mix. This is like taking a bunch of lottery balls and letting them fall 50% from each parent. Then taking half of those and mixing them with another 50% in the next mating and so on. The probability that the ball or genes will resemble anything from further generations is so great that it is meaningless. Sorry to say your program does not over come the mixing of the genes to out weigh the benefit of matching physical horses in the first two generations to be of any value. If it does can I adapt it to win the powerball jack pot

re: Announcing the All New Key Ancestors Report

sceptre — Tue, 15 Jan 2013 18:32:46 GMT

Pedigree Ann:

Yes, but there are populations, and then there are more "defined" populations-which is precisely the subject matter of this Report. For that matter, your example of Storm Cat vs Cee's Tizzy for Cee's Song is what this report strives to uncover...And what if you haven't yet found (for your mare) a mating that "works" (and even here there are degrees)? This Report might just aid in your process of discovering what "works".

re: Announcing the All New Key Ancestors Report

Ian Tapp — Tue, 15 Jan 2013 17:07:04 GMT

Sceptre,

Yes, all ancestors must meet a threshold of number of occurrences to receive an ancestor score. For a selected stallion (Bernardini, for example) to receive his own ancestor score, he must meet the same threshold (Bernardini would need to have been bred to enough of the related mares).

If Bernardini doesn't meet the threshold, then he would be quantified by the ancestor influences in his pedigree.

It doesn't matter if Bernardini happens to be the sire of a related mare, unless the related mare was a starter and her dam was also a related mare in the report.

re: Announcing the All New Key Ancestors Report

sceptre — Tue, 15 Jan 2013 16:43:54 GMT

Thanks, Ian; another very clear explanation.

Here's another question that may help clarify things:

Is it correct that an ancestor will receive no "value" (be accorded no weight) unless it meets a certain theshold number (of occurrances)? If that is correct, then should a "selected" stallion (proposed mate for the mare in question) have also sired offspring from the group of most similarly bred mares, this stallion, himself, would receive no "weight" (value) unless he occurred in a "sufficient" number of their offspring. What if this same stallion had also (for example) sired some of the most similarly bred mares?

re: Announcing the All New Key Ancestors Report

Ian Tapp — Tue, 15 Jan 2013 16:41:39 GMT

Hi Paul,

Related mares and their produce (the ancestors in the pedigrees of sires bred to related mares) are weighted only by their relatedness to the subject mare. There are no assumptions or measurements made specifically about inbreeding, since there won't be enough instances of each case of inbreeding to quantify it.

However, once we have the full list of ancestors and can divide them into elite and non-elite groups, we use a probability formula based on the population rate of elite runners, each ancestor's number of occurrences, and each ancestor's generational distance to determine which ancestors can be quantified within our confidence level. These are the ancestors whose names make it into the positive/negative ancestor tables in the Key Ancestors Report and are then evaluated per stallion in the Key Ancestors with Analysis.

In your Eclipse example, each occurrence of would count equally, weighted for its generation, but would not be counted if the occurrence is behind another quantified ancestor.

While we can't quantify the ideal level of inbreeding (at least not in this report), my opinion is that if you are going to mate your mare to create inbreeding to a particular ancestor, it would make sense to inbreed to a positive ancestor rather than a negative one.

re: Announcing the All New Key Ancestors Report

Pedigree Ann — Tue, 15 Jan 2013 15:45:19 GMT

As you folks know, I trained as a statistician (MS, PhD) so I am very aware that what is usual for a population as a whole and what is true for a particular member of that population may be quite different things. Storm Cat may have been the best stallion in the country at the time, but for Cee's Song (dam of Tiznow, Budroyale, etc.), Cee's Tizzy was a better match (check her produce record).

"Go with what works" is my first breeding maxim. If you find a mating that works for your particular mare, one that produces an outstanding individual, ignoring that datum is self-defeating.

re: Announcing the All New Key Ancestors Report

Paul — Tue, 15 Jan 2013 04:41:06 GMT

That is interesting. So you are using an inbreeding co-efficient of related mares and their produce and applying a possible success figure (elite group) to a potential sire? When you use the co-efficient are figures applied to the % of inbreeding co-efficient to various ancestors of the resulting foal? For example, how many times does Eclipse appear in a 16 cross peidgree and what % is that in the resulting foals genetic make-up? And would each of those ancestors have an equal genetic makeup figure based on it's place in the pedigree or would the results be weighted to the characteristics of an applied "elite" standard?

re: Announcing the All New Key Ancestors Report

Ian Tapp — Mon, 14 Jan 2013 16:29:18 GMT

Sceptre,

We assigned maximum and minimum values (+80.00, -80.00) to what an ancestor can contribute to a hypothetical mating in the Key Ancestors with Analysis. After running hundreds of reports and plotting the ancestor scores, we were able to determine these limits. Essentially these limits allow that any ancestor can be only so good or so bad.

Note that ancestor scores in the Key Ancestors Report with Analysis are weighted for their generational distance. If an ancestor has a score of +10.00, then he would be valued at +10.00 as the sire of the hypothetical mating, +5.00 in the 2nd generation, +2.50 in the 3rd generation, etc.

In some reports, there are certain ancestors that have not appeared in any elite runners despite the required opportunity to do so; these ancestors show up on reports as -Infinity (negative infinity). If there was no limit, any -Infinity ancestor would automatically disqualify any stallion carrying that ancestor, regardless of how distant the generation, which would be unrealistic.

To your 2nd point, this is a clever idea and something everyone would like to know: how repeatable are successful matings? This question comes down to probability, but due to environmental factors, the true probability of success is difficult to determine (You may have bred the next Man O'War, but he is struck by lightning in the middle of his first race). Also, the probability would change given the genetic material involved in each mating. I think Byron would be the person to talk more on this.

However, we have used a similar value in our Key Ancestors formula for how ancestors are evaluated. This is a proprietary value of the population rate of elite runners from starters (i.e. if ancestors are randomly distributed, what percentage of the ancestor's blood will appear in elite runners?).

re: Announcing the All New Key Ancestors Report

sceptre — Mon, 14 Jan 2013 05:21:20 GMT

Hi Ian,

Thanks for those clarifications, and I apologize for my sloppiness in not better reading the Report's instructions, and not analyzing more more carefully the math in the tables. It did, however, cause me to go back to the "drawing board"-re-read the Report's instructions, and review all charts and data contained in the Report. So, a few more questions and comments:

For a proposed mare, I notice the following Ancestor Scores: Fappiano +407.20; Caro +41.96. Going down now to the Hypothetical Mating Impact (for a selected stallion for the proposed mare) Fappiano (occurring solely in the 3rd generation) has a score of +20, while Caro (also occurring solely in the 3rd generation) has a score of +10.49. I'm sure there's a well thought out/reasonable and, no doubt, proprietary explanation for this-on the surface-mathematically comparative disparity, but I was wondering if you're able to shed any light?...On a somewhat separate note, I continue to be troubled (I suppose that's a fair description) by the possibility that one (single) close-up Stakes Winner could so distort the "numbers" as to drastically alter the "validity" (reality) of the Report for one or more influences (within that "line"). Since I'm sure you catch my drift, I'll jump far forward to some of my reasoning behind this thought: It is not uncommon for a stakes producing mating to be repeated. Somewhere in "data land" all such repeats, and their results (race record-wise) are documented and, therefore, likely can be subjected to mathematical analysis. Might not the end product of such analysis alter the "formula" you employ in your "weighting"?

re: Announcing the All New Key Ancestors Report

Ian Tapp — Mon, 14 Jan 2013 03:34:48 GMT

Sceptre, thanks for the supportive words about the report.

About the inbreeding—yes, that's an interesting aspect of the report. If the subject mare carries Seattle Slew, and the report finds that related mares did very well with sires carrying Seattle Slew, then the report has no issue with suggesting inbreeding to Seattle Slew. The acceptable level of inbreeding is of course up to the breeder, but if you're going to try an inbred mating, there should be some supportive reasoning behind doing it (not just because it looks cool on the five-cross).

Re: the Key Ancestor Score – this is in fact the net total of ALL ancestor scores (positives and negatives) in the sire's pedigree, not just the top 5 shown. The top 5 ancestors shown in the stallion summary tables are sorted by absolute value of their scores, giving weight to ancestors found in closer generations.

To your point about eliminating the subject mare's own produce from the scores: While it would be interesting for the reasons you mentioned, after all we are trying to find the best stallion choices for the subject mare, and, as you say, what better data to influence this than her own produce history.

One note—elite runners are not determined by earnings, which is generally not a reliable tool especially when you consider international runners (Japanese earnings, for example). Elite runners in this report are stakes winners and graded/group stakes-placed horses. While no threshold is perfect, the black type system is close.

re: Announcing the All New Key Ancestors Report

sceptre — Mon, 14 Jan 2013 00:36:53 GMT

I've now had the opportunity to take a more thorough look at this Report, and what it can provide to breeders. I have to say that much of my initial skepticism has been replaced with admiration for those that contributed to this project and eventual Report. It takes some time to fully appreciate what has been offered to us by this tool, but it's rather apparent to me that it will prove very useful in my matings-it appears to illuminate areas where even those most skilled and intense would often miss...A few more questions:

It appears that the Key Ancestor Score (for each selected stallion) is derived solely from the Top Five Ancestor's (by occurrance). I grant that this is a most valuable piece of information and should continue to be supplied. In addition, you also offer us tables for both the most positive and negative ancestors found in the produce of those most closely related mares (do I have that right?). Well, it occurs that there might be benefit in knowing what is the TOTAL Ancestor Score (derived from all ancestors in each selected stallion's pedigree). Would this not better illuminate the relative "value" of each stallion chosen for the mare in question?

I also see benefit to supplying, in a separate report, all data derived after ELIMINATION of the subject mare's produce, should there be any. While the data from her produce is vital in forming "predictions" (making selections) relative to future potential mates it presents two areas of concern:

1. Without the inclusion of her produce data, one might more easily ascertain whether or not she was appropriately mated in the past.

2. The subject mare's produce data could, in some instances, overly skew much of what is provided in the Report. Take, for example, a mare that had produced one $1M earner (by a particular stallion) and little else. (I assume amount of earnings factors into your "calculations" ?).