York Draw Bias: 5f, 6f and 7f (and Why the Sources Disagree)

Home › Racecourses › York Draw Bias

By The Guv'nor · Last updated October 2026

The York draw bias is low on the straight course (5f and 6f), and low means the far side. In 434 sprints of eight or more runners, 2016 to 2026, the bottom third of the draw won 180 (41%) against roughly 31% expected; the top third won 111 (26%). 7f is dead level. The mile and 10f favour low too.

Look, I'm essentially a statistics/maths geek, so I pulled the official result of every Flat race at York from May 2016 to the September 2026 meeting: 1,306 races. The tables use the 1,066 with eight or more runners (14,415 runners). It's one of the courses in my UK and Irish racecourse guide, and the draw bias guide has every other Flat track.

York draw bias by trip, 2016 to 2026

I split every field into thirds by stall after closing the gaps left by non-runners. Thirds rarely divide evenly, so a fair draw gives the low third about 31% of winners, the middle 34% and the high third 36%. That's the yardstick throughout.

5f, 6f, 1m and 10f lean low, and none of those looks like chance (chi-square test, p below 0.01). 7f, 14f and 2m are as fair as it gets. The 12f middle figure is borderline (p=0.04), and testing ten trips at once, luck alone throws up one of those.

Settling the argument: York sprint draw, high vs low vs middle

Which side is "low"?

On York's straight course, low is the far side. The official in-running comments show it: the 505 sprinters described as racing far side were drawn, on average, 29% of the way across the field; the 564 who raced stands' side sat at 82%. Where a winner's comment names a side (97 of 434 races), it's 52 far side, 13 centre and 32 stands' side.

Field size decides it

That's the heart of it. In a mid-sized field the draw barely matters (p=0.08). Once 14 or more line up, the low third wins half the races. It's the big-field sprint handicaps where the York draw really bites.

Why the sources disagree

One York season only gives you about 40 sprints of this size, and a sample that small swings about. Here's the straight course year by year:

Look at 2017, 2020 and 2025 alone and you'd swear the middle was best. The 6f races of 2023 went 11 of 23 to the high third, so a guide written that winter could easily say "high". Pool the lot and the low edge shows up in both halves of the period: 40% of sprints for the low third in 2016-2020 (187 races) and 43% in 2021-2026 (247).

The longest public study I found shows the same thing inside one article. Dave Renham on geegeez (April 2020, 8+ runner handicaps, 2009 to 2019) found a low bias over 5f that strengthened from 2015, and a 6f that looked fair over the long term but strongly low from 2014 (51 races). Over 7f he reckoned "a middle draw maybe optimal". Different years, filters and field sizes give different answers. That's all the disagreement is.

7f to 10f on the Knavesmire

York's own owners' guide describes the track as "left-handed, flat and fair", with a run-in of almost five furlongs (1,000 metres), and says 7f races start from a chute. At 7f the numbers agree: 56, 59 and 65 winners from 179 races (p=0.97), with nothing much changing in handicaps or on soft ground.

The mile is different. The low third won 48 of 107 races (45%), and 46 of 89 mile handicaps (52%). Over 10f, which covers the Dante and the Juddmonte International trip, low won 54 of 122 (44%) and high just 25 (20%). In 10f fields of 14 or more the high third won 6 of 53. Geegeez also found a low mile bias, putting it down to the short run to the bends.

From 12f up the draw tells you very little, as the table shows.

Going and rail movements

Only 279 of the 1,066 races (26%) were run on good to soft or softer. The sprint bias holds on both. On good or firmer, low won 133 of 327 straight-course races (41%) and high 89 (27%). On good to soft or softer, low won 47 of 107 (44%) and high only 22 (21%), so easy ground makes a high draw worse, not better.

The mile bias is clearer on quicker ground: low won 35 of 74 (47%) on good or firmer, against 13 of 33 (39%) on softer, which is too few races to call. Over 7f, high draws won 18 of 49 softer races, bang on their fair share, so I can't back up the idea (geegeez, from 28 races) that wide 7f draws struggle on easy ground. For what the descriptions mean, see my going explained guide.

The honest bit about rails: results don't record rail positions, so none of my numbers are split by them. On the day, check the clerk's update on York's going report page (the course points owners to @YorkClerk on X as well). One more trap: York's official round-course distances changed from the 2017 season. The mile is listed as 7f 192y, 10f as 1m 2f 56y and the Ebor trip as 1m 5f 188y, so match like with like when you compare old times.

Do punters over-react to the York draw?

If everyone knows low is good, the price should already reflect it. To test that, I compared each group's winners with the number the starting prices said they should have had (A/E: 1.00 means the market had it right) and worked out the return from £1 on every runner at SP.

So no, the market doesn't over-react. If anything it under-reacts: low-drawn horses win more often than their prices say, and high-drawn ones are overbet. Blindly backing low still mostly loses after the bookies' margin, and the 10f profit is hindsight that came from 2021-2026 (+34.8% on 288 runners, against -0.1% on 214 in 2016-2020). The draw earns its keep in steering you off short prices stuck on the wrong side.

How I use the York draw

Look, the draw is just another number to me, and a number only matters if the price hasn't already swallowed it. My game is profit month to month, not win rate, and I'd much rather find an outsider on the right side of a big-field sprint that can bolt up or at least place each-way than take a skinny price about a favourite drawn on the wrong side.

What the draw never changes is my staking. Same points whatever the stall (1pt = £5 on a £100 bank, as an example), no going all in because a horse is "perfectly drawn", and if I'm unsure, I skip the race. Even the best-drawn horse gets beaten more often than not. Understanding that losers are a thing is the key to being a winner longer term.

Dante and Ebor meeting notes

York has confirmed its 2027 fixtures: the Dante Festival runs Wednesday 12 to Friday 14 May (the Dante Stakes is on the Thursday) and the Sky Bet Ebor Festival runs Wednesday 18 to Saturday 21 August. Both are in my Flat season preview.

Dante meeting (May, 2016-2026; the 2020 Dante was run in July and isn't included): low won 32 of 71 straight-course races (45%) and high 15 (21%). At 10f and beyond the middle third won 31 of 59. That's suggestive, but it's a small sample.

Ebor meeting: low won 40 of 95 sprints (42%), middle 37 and high just 20 (21%). At 7f and 1m, low won 30 of 68 (44%), though that's borderline (p=0.05) and could be luck. From 10f up it was level: 32, 34 and 29 winners from 95 races. The October meeting's Sprint Trophy, a big-field 6f handicap, has gone to the low third in 8 of the 10 runnings from 2016 to 2025.

The Ebor itself is a heritage handicap over 1m 5f 188y; York calls it the richest Flat handicap in Europe at £500,000 (2026 figure). Its draw record is all over the place:

Closing the gaps for non-runners, that's six from the high third, two middle and three low. Eleven races prove nothing, and York's wider 14f record is level.

Want more on the draw at other tracks? Members can also read my Draw Biases In Flat Racing special.

York draw bias FAQs

What is the draw bias at York?

Low (far side) on the straight course, mostly in fields of 14 or more. 7f is fair; the mile and 10f favour low. That's from 8+ runner races, 2016 to 2026.

What is the York draw bias over 5f?

Low. The low third won 60 of 152 races (39%) and the high third 37 (24%), 8+ runners, 2016 to 2026.

Is high or low best over 6f at York?

Low: 102 wins (41%) for the low third from 246 races, 66 (27%) for high. Single seasons can say otherwise, as 2023 did.

Is there a draw bias at York over 7f?

Not one worth betting on. From 179 races the low, middle and high thirds won 56, 59 and 65, close to what you'd expect by chance.

Does the draw matter in the Ebor?

Not in any way I can measure. Winners since 2016 have come from stalls 1 to 24, and York's 14f races overall show no bias.

More in this guide

• UK and Irish racecourses: every track by direction, shape and draw
• Draw bias at every UK and Irish Flat course
• Chester draw bias, 5f to 1m
• Ascot draw bias and the Royal Ascot handicaps
• All-weather racing and the draw at every AW track