Understanding the most successful breeding horses in PFL

Breeding of horses in PFL is probably one of the most discussed but least understood topics. There is as much sophistry as there is science in the discussion with various theories doing the rounds.

• On track performance is genetic and not revealed in the attributes

• Gen 1 horses are superior to later generation derivatives

• Breeding is just a random process, the parents set the opening parameters but that’s it

• Some horses have superior blood

• There is a very limited amount of good quality studs

• Breeding outcomes can be predicted without understanding underlying genetic strings of numbers

This post is not attempting to dispel or confirm any of these belief systems but instead focus in on what we can see with our own eyes, which is the on track performance. We can’t see genetic potential, unique blood, or objectively measure bloodlines. What we can do is stick to some basic facts.

We've covered an intro to breeding before and this builds on that knowledge.

It takes two to tango

Which horse is the most successful parent, grandparent or great grandparent in the game? What does the stats section of PFL say? From a progeny earnings perspective it says Walter White.

This is a somewhat limiting piece of insight. What profit margin were those earnings generated at? What’s the average profit (not earnings) per foal? What is the win or podium rate of the average foal? We can’t calculate those things directly. Given the breeding costs might be 10k or 20k+ what is my expected racing ROI?

To try to answer some of these questions we need to summarise on track performance data of each horse and aggregate it up through the family tree so that we can see not just the direct offspring’s performance but any horse that carries the DNA of another horse, sire or dam.

Starting with a fairly factual assertion, let’s safely assume that 50% of the genetic material of a foal comes from each parent. Today the conversation about breeding is almost exclusively about the sire and studding. I feel a Michael Lewis / Moneyball moment / meme coming. If the focus is on studs, maybe mares are being slept on, and that’s where the value is.

The following analysis was conducted on performance data up to 14th June 2024, and bloodline data from 6th June 2024. I state these dates because already by the time of publishing the numbers will have changed, and by the time you find this article they would have changed again. So my advice would be to take these findings as directional rather than definitive; like a drunk uses a lamppost, for support not illumination.

Taking a snapshot of those horses whose progeny have an average win rate of at least 15% and an average winnings payout of 5,000 derby per race we narrow the field to 41 horses of which only 9 are male. The above scatter plot illustrates these 41 horses (please note the non-zero axes, usually the first trick in how to lie with statistics – but obviously not this time!)

K2 is a great example of why it is so important to consider not just the male blood line but the female blood line. Clearly, a premium mare is not going to be bred with the highest bidder, something that male studs have to contend with when looking at their offspring. If someone wants to breed a champion stud with their 3 start A- mare, you take the money and unfortunately the hit on your progeny stats. As a result this group is dominated by females. How many stables consider ways to get K2, Earthmother or Amumu? In fact at the time of writing Amumu, the +2 subgrade S broodmare is currently on the marketplace for a cool 4.2m derby. If a big spender wants to make a splash in the game, then why not start with one of the premium mares and have a pick of the studs?

There is a narrative doing the circles that many stable owners fast-bred their sub-par mares in an attempt to roll a high grade horse to immediately sell on the marketplace for a guaranteed profit. Well those days appear to be over. Foal prices, for any grade, have never been lower, and the surplus of studs on the market mean that there is excess demand with few studs selling out their 35 slots, regardless of what you might be told.

The reason fast breeding may not work is that the mares are sub-standard, and dilute the effect of having the blood of champions in the breed. Those stables that own premium mares are more mindful in who they breed with, and for those wishing to get ahead, acquiring or breeding these mares should be a priority.

Benedict, the highest derby winnings per race male, has sired Exoticus, Blue Norther and Caesar, despite winning just 1 race in 18 starts has an exceptional record through his progeny. Yet only 12 of his 35 breeding slots were used this season. Maybe there are more economic options below the circa 24k breeding price? Maybe the game has evolved. The performance of his foals and their foals has nonetheless been great.

A glimpse at the bottom twenty horses by win rate contains some cross-over low preference horses and some that have never bred a champion despite their own performance on the track. Amongst the columns, d_profit is a derby only calculation where dc_profit includes crown stimmies (converted to derby at 0.5$).

There are horses here, who despite their own on track performances are turning out offspring that struggle to win 5% of their races.

This is a good illustration of the random nature of breeding, Breeding Reports are simulations of 1,000 trial runs. You never know which one you are going to get. It’s quite profitable to predict breeding outcomes because no matter how far out you are, you can point to the stochastic nature of the process. Your foal will fall somewhere on the range of outcomes simulated on the breeding report. You can’t breed the same pair of horses 1,000 times to prove or disprove someone’s prediction. For good or bad you are stuck with the horse you receive.

The only reliable way, outside of hitting the jackpot, to beat a random process is through scale. If you have enough high end horses breeding with other high end horses of course you will invariably breed some champions. Then you can bask in the glory of your breeding strategy safe in the knowledge that is was brute force that won the day rather than deep understanding of genetics.

Walter White, the top grossing stud, has created 5 S+ grade horses out of 329 runners with a win rate of 15.5%. Morpheus’s offspring win at 19.3%, 1 in 3 are S+ grade (Morpheus is an S+ where WW is an S), and it costs 5k less to breed with him. There are many market inefficiencies, and breeding is one of them in this game. Look beyond the stats on display to find the value.

I don't want to spend time pointing out high cost studs with terrible on track records, but they are there, and its up to the canny stable owner to do their own homework before parting ways with their hard earned money. Regardless of how complicated breeding is implemented in the game, the only thing that matters is performance on the track, whether in horses racing now or future generations being planned out. If your horses don't race profitably on the track, I would say any insight into the genetics of breeding are probably wasted. If you are churning out as many bad horses as good ones, you are probably riding along on the tide of lady luck, and a prisoner to the random nature of breeding in the game. That doesn't mean we can't find and catch an edge though.

Measuring when it matters

Looking at all the foals, as previously mentioned includes a lot of misleading signals. The heart of racing is the graded stakes, from grade III to grade I. Let’s take a look at graded racing only and see which horses have the best record in the toughest races.

Again K2 demonstrates how dominance at the top of the table. K2 offspring (and their offspring) competed in 812 graded stakes races, winning 304 of them at a 37.4% strike rate. Amumu (available for just 4.2m derby on the marketplace at time of writing) is second place with a over 1 in 4 winners. The top male horse is once again Morpheus, followed by Midnight Rose and AquaEggclipse. For reasons already discussed, females dominate the list, despite the focus purely on graded stakes races.

Of the largest volume horses, the popular Shaun E Bear and Astro Nerd’s offspring win less than 10% of graded races entered, less than a third of K2’s win rate. The also have significantly negative ROI in these races when just considering derby (no stimmies). Does it make some of the most popular studs in the game bad? No, it just means that there are edges that are being left on the table.

What did we learn?

Earlier I asked the question “Which horse is the most successful parent, grandparent or great grandparent in the game?”

From this analysis at least, the answer is K2, not a 100k a breed stud, but a broodmare. Maybe its more important to figure out how to get some of her blood into your breeding plans than the pick of the month stud that you keep seeing advertised. Getting hold of high value mares is very difficult, but as mentioned already the super impressive Amumu is currently (for how long?) listed for 4.2m derby. My finders fee is 10% so if you do end up buying her remember to send me the cheque in the post.

When planning bloodlines, mares are just as if not more important than studs. They are harder to acquire, they only breed once a season, you can’t buy those breeds on an open market, and the choice mares were snagged very early in the game’s lifecycle, making it difficult for newer stables to get involved. That said, hunting down progeny with blood that has been passed down should be your plan. Keep your eyes open for the horses mentioned here in the bloodline charts of horses you are reviewing. Ask around, use community tools and alerts like PFE or Gap or MJ.

Join the fun and put these insights into practice at PhotoFinish.Live and if you are considering starting your own stable please consider using my referral code: PADDOCK or just click on this link: https://signup.photofinish.live/?referralCode=PADDOCK

Please remember this is a web3 game where your spend your own money. Nothing I write about should be considered financial or investment advice.

Other blog posts:

• What are subgrades: https://aipaddock.com/what-are-subgrades-and-how-do-they-work

• Fastest horses: https://aipaddock.com/who-are-the-fastest-horses-in-the-game

• Racing profitability: https://aipaddock.com/how-much-profit-do-you-make-racing-horses

• Trueskill pvp ratings: https://aipaddock.com/which-horse-is-the-goat-in-pfl

• Breeding: https://aipaddock.com/trying-to-understand-breeding-in-pfl

• Evaluating horses: https://aipaddock.com/how-good-is-my-horse

• Understanding performance: https://aipaddock.com/understanding-the-true-performance-of-your-horse

• Do stars matter: https://aipaddock.com/how-much-do-preferences-matter

• FF Rating vs Finish Time: https://aipaddock.com/the-difference-between-ff-rating-and-finishing-time

• Are horses getting faster: https://aipaddock.com/are-pfl-horses-getting-faster