Pedigree, conformation, physical scope, sale performance, market demand and vet risk. A lot is given the ones its kind of sale can be asked. Conformation and vet risk are not offered today. Each read carries its own evidence and its own gaps. This page shows you both.
The reads
The horse on one side. The price on the other.
A single number can quietly bake the price in. Ours cannot, because the reads are worked out separately and shown separately. The status beside each one is counted off the sale, not written down here.
What the horse is
Pedigree
What the family says about this horse's chances.
Not yet scored
Conformation
How the horse is put together.
Not offered today
Physical scope
Size, frame, and how far along the horse is for its age.
Not yet scored
Sale performance
What the horse did in front of buyers on the sale ground.
Not at these sales
Vet risk
Whether anything a vet finds should worry you.
Not offered today
Price stops here
What the market will pay
Market demand
What comparable horses by the same sire have actually sold for.
In use today588 of 768 lots · high and low confidence
Fold price into ability and a cheap horse starts to look talented for being cheap, and a dear one looks talented for being dear. So the reads never touch. What the market will pay can tell you a horse is expensive. It can never tell you the horse is slow.
Where the reads come from
What each read is built from, today.
Not what it will be built from. Some are still waiting on evidence that does not reach us yet, and every count below is taken off the 768 lots standing in the two Karaka 2026 books.
Pedigree
Not yet scored
What the family says about this horse's chances.
In hand today
Nothing. This read has no evidence at all yet, which is why no lot carries it.
Still missing
How often the sire's runners win
How many of them reach stakes level
The dam's own race record
Black type through the family
How the sire and the damsire's line have crossed before
Those all come from the official racing record, and we do not have access to it yet. Sale prices tell you what the market thought of a stallion, not what his runners did on the track, so they are kept out of this read, and it stays blank without the record.
No lot of the 768 standing in these books carries this read today.
Conformation
Not offered today
How the horse is put together.
In hand today
Nothing. This read is not offered today.
We do not analyse conformation today. Vendors take stand-up photographs, but nothing in the app reads them, so this read stays blank rather than borrowing an answer from the pedigree, which is a different question about a different thing.
We do not offer this read today, so no lot in these books carries it.
Physical scope
Not yet scored
Size, frame, and how far along the horse is for its age.
In hand today
Nothing. This read has no evidence at all yet, which is why no lot carries it.
Still missing
Measurements taken at the sale ground
How far along the horse is for its age
Physical scope comes from a completed inspection. None is recorded against these lots, so the read shows nothing at all.
No lot of the 768 standing in these books carries this read today.
Sale performance
Not at these sales
What the horse did in front of buyers on the sale ground.
In hand today
Nothing, and nothing is expected here at these sales.
Still missing
The published time from a sale where the horses work in front of buyers
The sale's own record of what each horse was asked to do
At a yearling sale nothing is asked of the horse in front of buyers, so this read does not appear on a yearling lot at all. It is not waiting for anything: there is nothing to wait for. At a sale where the horses gallop a short breeze and are timed, the time is what this read is built on, and it is the one piece of evidence here we have measured against what horses actually made. The Ready to Run catalogue in the app has no published times yet, so the read is empty today.
This read does not apply to the kind of sale these books are, so no lot in them is given it.
Market demand
In use today
What comparable horses by the same sire have actually sold for.
In hand today
Sale results for lots by the same sire, from this sale or from past sales of the same kind
Still missing
A close enough comparison for every lot
Shown as a range, never a single number, and the range carries its own confidence. Where no comparable set can be built out of real results, no range appears at all. The gap is the honest answer, not a broken screen.
588 of the 768 lots standing in these books carry this read today. Of those, 513 at high and 75 at low confidence.
Vet risk
Not offered today
Whether anything a vet finds should worry you.
In hand today
Nothing. This read is not offered today.
We do not analyse x-rays or give a veterinary read today. No vet reads films for us, so this read stays blank, because we will not call a horse sound on data we do not have. This is decision support, not veterinary advice, and it never replaces your own vet.
We do not offer this read today, so no lot in these books carries it.
What we have measured
Quicker horses made more, and not just because of the sire.
At a sale where the horses gallop a short breeze in front of buyers, they are timed. Those times were measured against what the horses actually made, at five sales in a row from 2021 to 2025. Then we took out everything the market was already paying for the sire, and measured what was left. The link held at every sale.
Each sale was measured on the horses that carried a published time, sold, and had another horse by the same sire at that sale to set the sire’s standing against.
20210.37154 horses
20220.26166 horses
20230.39202 horses
20240.36219 horses
20250.40235 horses
The figure is how closely the two ran together once the sire was set aside, on a scale where 0 is no link at all and 1 is a perfect one. If there were no real link, a pattern this strong would turn up by chance less than once in a thousand times, at each of the five sales separately. That says the pattern is real. It says nothing about how well it reads any one horse.
Faster breeze times went with higher prices at every sale we looked at, and kept going with them after the sire's standing was set aside. That is what was measured and all that was measured. It does not say what any one horse will make, and we do not use it to say so.
The times only compare inside one sale.
Nothing on record says how far the horses breezed. So a 10.4 at one sale and a 10.4 at another are not the same test, and the sales must never be pooled into one number. Everything the app shows is ranked inside a single sale for that reason: you see where a horse sat among the horses it ran against, never a figure you could carry to another sale.
The measurement is drawn from the horses that sold.
A lot that does not sell has no price, so it cannot be in a measurement of times against prices. The horses that failed to sell are missing from every sale here, and they are not a random sample of the catalogue: the slower ones are likelier to be among them. That makes the real link a little different from the measured one, and we cannot say by how much.
It is one kind of sale, and one clue on the day.
Only at a sale where the horses breeze is there a time to measure, so this says nothing about a yearling, and nothing about a horse whose time was not published. On the horses it does cover it is one clue among the others, and it is read as one.
What the range is worth
Our ranges mean what they say. About half of horses land outside them.
Where there is enough to go on, a lot is given a price range built from what comparable horses have made. We checked it the only way that counts: on a sale that had already happened, with every range built from the earlier sales alone, then held against what each horse made on the day.
50 of 99
horses made a price inside the range built for them.
Measured on 99 horses at the 2025 sale: the ones with enough comparable sales behind them to be given a range at all, that then sold. The ranges were built from the 2021 to 2024 sales, none of which included the horse being checked.
25 made less than the bottom of their range and 24 made more than the top. The misses fell almost evenly either side, which is a range sitting where it should rather than high or low. It is also about half of horses landing outside the range in front of them, which is the honest shape of a range and not a fault in it.
A range is not a price.
It says where comparable horses have landed. It does not say what this one will make, and about half of them make something outside it. A horse can go well past the top of its range or fall under the bottom, and neither of those is a surprise.
The check is drawn from the horses that sold.
A lot that does not sell has no price, so there is nothing to hold its range against. Those horses are missing from the check, and they are not a random sample: a lot passes in where the seller and the room disagreed most.
One sale, and the ranges only reach some of a catalogue.
The check is one Ready to Run sale, with the ranges built from the four before it. It is not a yearling result and it is not a promise about the next sale. A lot with too few comparable sales behind it gets no range at all rather than a thin one, so on a real catalogue a good many lots show nothing here.
What we will not do
A blank is a decision, not a bug.
A page that admits what it cannot see is worth more at the sale ground than one that fills every field.
We will not call a horse sound on data we do not have.
We do not analyse x-rays or give a veterinary read today. Silence is the honest answer and it is the one you get.
A read with no evidence says Not yet scored.
Never a zero, never a dash, never a plausible-looking number standing in for one we do not hold.
Price never moves a read on the horse.
And a read on the horse never moves the price. The two are worked out apart and shown apart.
On the lot itself
The same rules, on a real horse.
A real lot out of the sale, with the reads the app holds for it today. What the engine holds for this horse is shown. What it does not hold is named rather than filled in.
The readA real lot from Karaka 2026 Book 1
Lot 292Satono Aladdin x Avistobay filly · Curraghmore
Value
backed by comparable sales
$90k-$250k
Pedigree
What the family says about this horse's chances.
Not yet scored
Conformation
How the horse is put together.
Not offered today
Physical scope
Size, frame, and how far along the horse is for its age.
Not yet scored
Vet risk
Whether anything a vet finds should worry you.
Not offered today
We will not call a horse sound on data we do not have.This is what the engine says about this horse today. The value read is built from real comparable sales. Anything marked Not yet scored is blank because the data behind it has not arrived, and we would rather show you that than fill it in. Conformation and vet risk are not offered today.Reads fill in per horse only when the evidence behind them reaches us. What is missing stays named rather than guessed.Open this horse in full
Bloodstock AI is built in New Zealand for the people the sales were never designed for: first-time buyers, small syndicates, and the agents who look after them. We put the reasons on the table and leave the decision with you.
