Book 4 Changed the Shape of Keeneland September

I hadn't planned to write much about Book 4. I assumed it would simply extend what we had already seen at Keeneland this September and follow the usual Book 4 arc. The numbers turned out to be more interesting than I thought.

Through the first three books, higher prices had generally come with weaker initial clearance. Book 4 reversed that relationship. Keeneland offered almost exactly the same number of horses as last year, yet sold 27 more of them, generated another $8.6 million in gross and lifted both average and median.

The way that additional money entered the market is even more interesting. Fillies generated most of the year-over-year increase. The middle 50% of the sold population appreciated faster than the Top 10%. Major buyers accounted for a smaller share of spending, while money became somewhat more concentrated around favored sires. And for the pinhookers who bought these horses as weanlings last November, a stronger Book 4 produced a much more mixed economic result.

Book 4 Improved Price and Clearance at the Same Time

The broad numbers are straightforward.

Figure 1 · The reversal

Price and clearance moved the same direction

Show the data

Book 4	2025	2026	Change
Sold through ring	555	582	+27
Gross	$54.60M	$63.16M	+15.7%
Average	$98,382	$108,516	+10.3%
Median	$75,000	$85,000	+13.3%
Through-ring clearance	75.3%	79.2%	+3.9 pts.

Only two fewer horses went through the ring this year. Buyers completed 27 additional transactions and spent $8.55 million more.

In Books 1 through 3, prices rose while initial RNA rates also moved higher. In Book 4, buyers paid more and bought a larger share of what was offered.

The longer view makes the move even clearer. Book 4 generated roughly $46.1 million in both 2023 and 2024. Gross jumped to $54.6 million last year and $63.2 million this year. The average has progressed from roughly $76,000 in 2023 to $80,000, $98,000 and now $109,000.

Through eight sessions, Keeneland has generated about $493.2 million from 1,732 through-ring sales, compared with $472.2 million from 1,771 sales at the same point last year. Book 4 alone supplied roughly 41% of that $21 million increase in cumulative gross.

For a part of the sale that accounts for less than 13% of current gross, that is a substantial contribution.

The Middle Had the Better Year

The session toppers naturally attracted attention, particularly the two Yaupon colts that brought $825,000 and $850,000. Together, however, they represented less than 3% of Book 4 gross.

The broader distribution tells us more about what happened.

Figure 2 · The middle

The middle of the market appreciated fastest

Show the data

Price-ranked segment	2025 average	2026 average	Change
Bottom 25%	$22,696	$24,679	+8.7%
Middle 50%	$76,960	$87,845	+14.1%
Top 10%	$289,643	$311,780	+7.6%

The middle 50% appreciated faster than either end of the market. Its share of gross increased from about 39% to 40.5%, while the Top 10% share slipped slightly.

Fixed price levels tell the same story. Last year's Book 4 produced 201 horses at $100,000 or more. This year there were 256, an increase of 27%. At $200,000 or more, the count rose from 63 to 87.

Six-figure sales therefore represented 44% of all through-ring transactions in Book 4, compared with 36% last year.

For breeders selling the kind of commercial yearling that typically trades in the $100,000 to $200,000 range, that may be more useful than another record session topper. The number of completed transactions at those prices expanded considerably.

The bottom moved too, although much less dramatically. The price separating the lowest 10% of sales from the rest of the book remained at $20,000. Most of the improvement came through the middle.

Most of the Extra Money Went to Fillies

This was the result I found most surprising.

Figure 3 · Where the money went

Fillies supplied nearly four fifths of the gain

Show the data

Book 4	2025 colts	2026 colts	2025 fillies	2026 fillies
Offered	376	379	361	356
Sold	297	305	258	277
Average	$106,488	$109,856	$89,050	$107,042
Median	$80,000	$80,000	$70,000	$85,000
Clearance	79.0%	80.5%	71.5%	77.8%
Gross	$31.63M	$33.51M	$22.98M	$29.65M

Colt gross increased by about $1.9 million, while filly gross increased by $6.7 million. Fillies therefore accounted for approximately 78% of Book 4's entire year-over-year increase in gross.

Their average rose 20%, compared with 3% for colts. Their median increased from $70,000 to $85,000, while the colt median stayed at $80,000. Buyers also took home 19 more fillies despite having five fewer offered.

The colt-filly gap had already narrowed in Book 3. Book 4 carried that trend farther, with the median sold filly actually bringing more than the median colt.

The sale results alone cannot tell me why. Broodmare residual value, the quality of this particular group and differences in reserve expectations could all contribute. What the data do show is where the incremental money went, and in Book 4 it went overwhelmingly toward fillies.

Yaupon Was Much More Than Two Big Horses

Yaupon supplied both Book 4 session toppers, which made him the obvious sire story. His complete group is more interesting than the two expensive horses.

Yaupon generated 57% more Book 4 gross with six fewer horses offered. His median more than doubled and 18 of 20 found buyers through the ring.

The $825,000 and $850,000 colts obviously affect the average, so I removed them. The other 16 sold Yaupons still averaged $193,250, with a $200,000 median.

Half of the entire sold group brought at least $200,000 in Book 4.

That follows the pattern Yaupon has shown throughout September. More supply has reached the market, and buyers continue to support the group well beyond the obvious horses at the top.

Life Is Good Passed a Different Test

Life Is Good's Book 3 numbers caught my attention because his second-crop prices moved sharply higher. Book 4 tested something else: how much supply buyers would absorb at roughly the same price level.

Last year, eight Life Is Goods were offered in Book 4 and six sold. They averaged $185,333 with a $190,000 median.

This year, 15 were offered and 14 sold, an initial clearance rate of 93.3%. They averaged $212,786 with a $192,500 median.

The median barely changed, while the number sold more than doubled. A stallion's market can strengthen through higher prices or through buyers absorbing more horses at an established price. Life Is Good has now shown both during this sale.

Through the first four books, 62 of 68 Life Is Good yearlings have sold through the ring, a clearance rate above 91%, with a $350,000 median.

Yaupon and Life Is Good together supplied about 42% of Book 4's increase in gross. Even after removing both sire groups, the rest of the book still produced higher average and median prices along with better clearance.

Their contribution was substantial, but the improvement extended well beyond them.

Some Earlier Sire Reads Changed

One reason Book 4 is worth studying is that catalog position changes the commercial question.

At the front of the sale, buyers and sellers are negotiating over horses expected to bring substantial prices. By Book 4, the same sire can meet a different group of physicals, pedigrees and reserve expectations.

Mage is the clearest example.

Through Books 1 through 3, only 12 of 26 Mage yearlings sold through the ring, leaving an initial RNA rate of 53.8%. The median on those successful sales was $140,000.

Book 4 produced almost the reverse result. Twelve of 14 sold, an 85.7% clearance rate, with an $80,000 median.

The earlier books showed substantial friction between buyers and sellers. Once the group reached a lower price point in Book 4, transactions became much easier to complete. Different horses make up the populations, so price alone cannot explain the change, but Mage clearly found a more liquid market deeper in the catalog.

Munnings showed a similar pattern in a same-book comparison. Last year, six of 12 sold in Book 4 at an $85,333 average and $70,000 median. This year, 11 of 12 sold, averaging $63,182 with a $55,000 median.

The price level came down and the market cleared almost the entire group. That adds useful context to the weakening clearance I had seen earlier in the sale. Buyers and sellers found agreement much more readily at a lower valuation.

Epicenter moved in the opposite direction.

His average fell 58%, the median fell 40%, and only half of the offered horses sold through the ring.

The Book 4 result makes the Epicenter weakness harder to dismiss as ordinary second-crop softening. His earlier yearlings had already come back in price, and this group brought weaker prices and weaker liquidity at the same time.

Maxfield provides a more moderate reminder that sire trends rarely move in a straight line. His first three books showed a substantial rebound from last year. In Book 4, five of seven sold for an $85,400 average, down from $102,500 last year, although the median increased from $75,000 to $85,000.

His September remains improved at the top. Book 4 makes the broader picture less uniform.

The Back Half Is Starting to Test the Freshmen

After Book 3, I wrote that the front half of September would tell us which stallions could produce an expensive yearling. The later books would tell us more about their commercial floor.

Book 4 started answering that question.

Figure 4 · The freshmen, deeper in the book

Book 4 tested the floor, not the ceiling

Show the data

First-yearling sire	Offered	Sold	Initial RNA rate	Average	Median
Gunite	32	24	25.0%	$131,375	$115,000
Taiba	26	22	15.4%	$116,341	$100,000
Forte	19	15	21.1%	$131,333	$100,000
Up to the Mark	22	18	18.2%	$98,556	$80,000
Mage	14	12	14.3%	$100,583	$80,000
Arabian Lion	14	11	21.4%	$79,818	$75,000
Arcangelo	17	13	23.5%	$80,769	$80,000
Elite Power	21	16	23.8%	$61,313	$43,500
Cody's Wish	7	3	57.1%	$151,000	$160,000
Annapolis	5	5	0.0%	$120,000	$125,000

Gunite and Taiba continue to look very commercial deeper into the sale.

Gunite handled the largest group in the table and maintained a $115,000 median. Taiba cleared 22 of 26 with a $100,000 median and the lowest RNA rate among the larger samples.

Forte remained consistent at a $100,000 median with 15 of 19 sold. Up to the Mark continued to clear well, selling 18 of 22, although his $80,000 Book 4 median sits below Gunite, Taiba and Forte.

Up to the Mark's top-end and middle-market performance remain strong, while Gunite and Taiba are showing particularly good support deeper in the catalog.

Arabian Lion also changed slightly. His 12.5% initial RNA rate through Book 3 had been one of the cleanest profiles in the class. He sold 11 of 14 in Book 4, still a reasonable result, though much closer to the overall market.

Elite Power sold 16 of 21, which is solid enough liquidity. The median was only $43,500. His September market increasingly looks like one with a valuable upper end and a much lower floor underneath it.

Cody's Wish remains unusual. Three of seven sold through the ring, but those three carried a $160,000 median. The successful horses can still bring substantial money while the group continues to encounter more resistance than most of the other leading freshmen.

Annapolis is now nine for nine through four books. The population remains small, but five Book 4 sales at a $125,000 median add another useful data point.

More Buyers, Narrower Sire Preferences

Book 3 produced an interesting combination: money became less concentrated by sire and more concentrated among the biggest buyer labels.

Book 4 moved in the other direction.

Buyer labels are as reported. Partnerships, agents and related ownership groups can appear under separate names, so this measures reported purchasing concentration rather than consolidated beneficial ownership.

The Top 10 buyer labels accounted for a smaller share of Book 4 gross than last year, while the number of reported purchasing labels increased. At the same time, the Top 10 sires captured a larger share of the money.

Book 4 therefore had a broader buyer bench directing more of its spending toward a somewhat narrower group of stallions.

The market appears less dependent on a handful of buyers at this price level, while sire preference remains highly relevant.

Buyer labels come with an obvious limitation. Partnerships, agents and related ownership groups can appear under separate names, so this measures reported purchasing concentration rather than consolidated beneficial ownership.

A Better Sale Was a Tougher Pinhook

The matched pinhook data provide a useful check on all the strong sale statistics.

I matched Keeneland November 2025 weanlings to their Book 4 September appearances by sire, dam, sex and date of birth. Eighty-seven assumed pinhooks were offered in Book 4 after excluding another 15 matched horses that were withdrawn.

Sixty-seven sold through the ring. Five more subsequently sold privately, while 15 remain recorded as RNAs.

For the 67 successful ring resales:

A 1.25x median gross multiple leaves very little room once the cost of carrying a weanling for ten months enters the equation.

Even before commissions, board, mortality insurance, veterinary work, sales preparation and financing, 26 of the 67 successful ring resales brought less than their original November purchase price.

That is 39%.

Book 3 looked quite different. The matched successful resales there produced a median gross multiple of about 1.73x, and only 10 of 56 brought less than their November purchase price.

Book 4 was a better auction for completing a yearling sale. It was a much tighter trade for many of the people who bought those horses as weanlings.

There were some excellent individual pinhooks. A Corniche filly bought for $70,000 in November brought $375,000. A Cody's Wish filly went from $50,000 to $285,000. A Taiba filly moved from $70,000 to $210,000.

Those trades sit beside 26 successful resales that brought less than their November purchase price, plus another 20 horses that initially failed to sell or required a later private transaction.

Rising yearling averages describe what buyers are paying now. Pinhook returns depend just as heavily on what sellers paid to enter the trade.

The RNA Floor Firmed Too

One final number helps complete the Book 4 picture.

Among the horses still recorded as RNAs, the median reported bid was about $55,000 last year. This year it is approximately $65,000.

Book 3 gave a different result, with the median unresolved RNA bid remaining around $95,000 year over year even as sold prices increased. Book 4 shows some movement underneath the unsuccessful horses as well.

An RNA bid needs careful interpretation because the reported amount does not establish how much independent bidding occurred or how far the reserve sat above it. Paired with the improvement in initial clearance, however, the higher reported bid level is consistent with a firmer market through this part of the catalog.

Where Book 4 Leaves the Sale

Book 4 changed my view of this year's September sale more than I expected. The first three books showed extraordinary purchasing power alongside increasing selectivity. Book 4 showed that demand can broaden when the horses, prices and seller expectations meet at the right level.

The most important change was beneath the headlines. The middle of the market strengthened, fillies absorbed a disproportionate share of the additional money, and clearance improved at the same time. Several sire stories also became clearer as the catalog moved deeper.

The pinhook data provide the counterweight. A stronger Book 4 yearling market still left 41% of the successful matched resales below their November purchase price before expenses.

That may be the most useful takeaway after eight sessions: the market for the yearling can be improving while the economics for the person selling it get tighter.

Thoroughbred Intelligence analyzed Keeneland September Book 4 results at the hip level, with year-over-year comparisons to 2025 and longer-term macro comparisons where available. Offered figures exclude withdrawals. Sold, average, median and gross use through-ring transactions unless otherwise indicated. Initial RNA rates retain later private sales as RNAs so that clearance reflects the result when the horse passed through the ring. Buyer concentration uses reported buyer labels and does not attempt to consolidate related ownership groups or agents. Pinhooks are matched by sire, dam, sex and date of birth and assume continuous ownership between the Keeneland November 2025 purchase and September 2026 resale. Gross pinhook multiples exclude commissions, insurance, board, veterinary expense, sales preparation, financing and other carrying costs.