THE DRAW IN UK FLAT RACING – A BROAD OVERVIEW (PART 2)

 

By the time you read this I wonder how close we will be to seeing live horse racing again in this country? I researched and wrote this piece back in April during lockdown and it looked a long way off then. Fingers crossed ……

 

Moving on….. in my last piece I had a generic look at the draw across UK courses at 5 and 6 furlongs. The focus was on 10 or more runner handicaps and for this ‘follow-up’ I am going to extend the research to qualifying races over 7 furlongs and 1 mile. Over 5 and 6f there are still a few playable draw biases so I wonder what we will uncover this time around?

 

As with the previous article I have taken data from 1st January 2015 to 31st December 2019 focusing on UK turf flat racing only. Once again I am splitting the draw into halves – comparing the bottom half of the draw (low) with the top half of the draw (high). A/E values (index) will be shared too. I explained in detail what A/E values were last time – in simple terms they are a type of impact value. The higher the figure the better in terms of potential value / profitability. From these initial findings I will drill down deeper hoping to find some useful nuggets of information.

 

7f handicaps – so let me look at 7f handicaps first. I have considered only those courses that have had at least 25 qualifying races. The table will show the win percentages for the both halves of the draw as well as their respective A/E indices.

 

Track	low ½ of draw (%)	High ½ of draw (%)	A/E   (low ½  of draw)	A/E   (high ½  of draw)
Ascot	52.78	47.22	0.80	0.72
Ayr	61.64	38.36	0.97	0.70
Beverley	49.02	50.98	0.70	0.99
Brighton	54.05	45.95	0.85	0.81
Catterick	44.62	55.38	0.68	1.06
Chepstow	26.67	73.33	0.44	1.20
Chester	66.67	33.33	0.85	0.79
Doncaster	47.37	52.63	0.75	0.87
Goodwood	82.50	17.50	1.08	0.38
Leicester	47.54	52.46	0.69	1.02
Lingfield	36.59	63.41	0.59	0.99
Musselburgh	51.79	48.21	0.82	0.93
Newbury	38.46	61.54	0.56	1.20
Newmarket (July)	35.90	64.10	0.56	1.10
Newmarket (Rowley)	51.28	48.72	0.81	0.84
Redcar	75.00	25.00	1.06	0.50
Salisbury	51.72	48.28	0.81	0.87
Thirsk	34.88	65.12	0.58	1.09
Yarmouth	46.88	53.13	0.71	0.99
York	51.22	48.78	0.77	0.82

 

There are a few courses that suggest there is a draw bias and I am going to start with Goodwood. The 7f trip at Goodwood is run round a bend and lower draws start closest to the inside rail. Hence assuming they jump well from the gate, lower draws should be able to take advantage of the inside, which of course is the shortest route. In addition to this the ground closest to the rail in the straight has traditionally been slightly quicker than the rest of the track, accentuating this potential edge for lower draws. The bias was very strong back in the late 90s / early 2000s and course officials at Goodwood eventually became aware of this. From around 2006 they started to negate the bias somewhat by moving the rail at times and I suspect also by some extra watering of certain parts of the track. However, I have noticed the bias has returned in more recent times and the stats in the table above back this up. In terms of the win percentages, the bottom half of the draw have won nearly 5 times as often as the top half. Looking solely at wins it is 33 to 7. Breaking it down further numerically, draws 1 to 6 have won 30 races from 240 runners (SR 12.5%) with a decent A/E value of 1.09. Backing all runners ‘blind’ would amazingly have seen you in profit by around 10p in the £ at SP; this rises to over 40p in the £ if you had used Betfair SP. Draws 7 and higher on the other hand have produced only 10 winners from 306 runners (SR 3.3%) with a poor A/E value of 0.46. Backing all these runners ‘blind’ would have lost you 64p in the £ at SP; 53p in the £ at BSP. Goodwood over 7f are races we need to check out in the future.

 

Redcar race on a straight track over 7 furlongs. The basic stats from the table indicate a reasonably strong low draw bias at this distance. Both the win percentages and the A/E values indicate this. Over 6 furlongs if we look back at the stats in the previous article, they too indicated that lower draws looked best there too – the potential bias was not as strong but still it seemed a relatively significant edge for lower draws. Digging more deeply into the Redcar 7f stats, the place data strongly favours low draws too which is good news. The bias that the place data indicates is actually stronger than Goodwood’s place data bias. Having said all that I still have reservations about whether this is a definite bias or not. I am hoping that when I look at the 1 mile stats later, they too show a low draw bias. The mile trip is also run on the straight course at Redcar and any positive correlation there would massively increase my confidence.

 

Chester has two 7f trips (7f and 7.5f) and not surprisingly sees a strong edge in terms of percentage wins for the lower part of the draw. However, the A/E values are similar and as with the 5f data I discussed last time the market is all too aware of the bias and prices are balanced accordingly. For the record, you would have made a decent profit backing the two highest draws despite a relatively modest strike rate. Sometimes it is worth going against the ‘grain’.

 

The last course to discuss before I move onto mile handicaps is Thirsk. At Thirsk they race left handed over 7f but it seems that lower draws nearest to the inside rail are at a disadvantage. This may be because there is quite a long way to the first and only bend, potentially helping to negate an inside draw. An interesting ‘double’ stat to share is this – the two lowest drawn runners have combined to win just 1 race in 86; the two highest draw runners have won 16 races from 86 runners. There is potentially some value in higher drawn horses over this trip at this Northern track.

 

1 mile handicaps – time to look at 1 mile handicaps now. Again 25 races per course is the minimum number I have chosen to qualify:

 

Track	low ½ of draw (%)	High ½ of draw (%)	A/E   (low ½  of draw)	A/E   (high ½  of draw)
Ascot	54.55	45.45	0.79	0.73
Ayr	62.26	37.74	0.98	0.69
Bath	38.89	61.11	0.62	1.06
Beverley	58.06	41.94	0.83	0.80
Brighton	48.98	51.02	0.81	0.86
Carlisle	52.94	47.06	0.85	0.83
Doncaster	62.96	37.04	1.00	0.61
Hamilton	65.85	34.15	0.97	0.67
Haydock	54.35	45.65	0.81	0.85
Leicester	57.69	42.31	0.86	0.81
Newbury	53.57	46.43	0.78	0.87
Newmarket (July)	52.00	48.00	0.87	0.80
Newmarket (Rowley)	58.82	41.18	0.96	0.71
Nottingham	61.54	38.46	0.92	0.74
Pontefract	65.85	34.15	0.85	0.82
Redcar	48.08	51.92	0.72	1.00
Ripon	48.48	51.52	0.70	1.03
Sandown	54.55	48.48	0.83	0.89
Thirsk	46.00	56.00	0.78	0.91
Windsor	59.62	42.31	0.88	0.78
York	70.97	29.03	0.99	0.53

 

First course to discuss is Redcar and it is disappointing to note that there is no low draw bias. The stats are very even. Earlier in the article it seemed that the North Yorkshire course had a decent looking edge for lower drawn horses over 7f, but without the correlation here I am less convinced there is actually a bias. The figures may simply be down to ‘chance’.

 

On a more positive note, York over 1 mile has always been a C & D that have given low draws an edge. The recent data once again indicates this continues to be the case. If we focus in on draws 1 to 6, they have provided 22 wins from 186 runners (SR 11.8%) and would have provided a ‘blind’ profit backing all runners – 12p in the £1 if using traditional SP and 26p in the £ at BSP. Horses drawn 7 or higher have won just 9 races from 291 (SR 3.1%) creating losses of 44p in the £ at SP (20p in the £ at BSP). Very high draws have a dreadful record at York over a mile – the three highest draws have combined to win just 1 race from 93 attempts!

 

Pontefract, like York is another course that traditionally has seen lower draws have the advantage over 1 mile. The initial win percentages indicate this but the A/E values are quite even. However, I have investigated further and the low draw bias not only looks strong but there is potential value there too. We need to focus in on the very lowest draws – horses drawn in the bottom quartile (1/4) actually have an A/E value of 1.04 and concentrating on the two lowest draws only, I can report that they have won 14 of the 41 races which equates to over 1 in 3 of all races. The low draw bias seems to increase on softer ground too – on good to soft or softer the A/E value for the bottom quartile of the draw increases to 1.26.

 

Ayr over 1 mile is another course and distance that has offered a draw edge in the past to lower drawn runners. The raw stats for both the win percentages and the A/E values seem to back that up. Digging a bit deeper we can see that the bottom three draws have won a third of all races with an A/E value of 1.06. Also they would have made a small profit of 14p in the £ to SP by backing them all ‘blind’ (30p in the £ to BSP).

 

Clearly draw biases do still exist – yes there are fewer than there used to be, and generally the biases are less strong. However, having a good appreciation of draw bias is still important when evaluating certain races. Any edge you can find is clearly important.

 

One caveat to the article that I must add is that although I have highlighted certain draws, or groups of draws have proved profitable in this recent 5 year period, I would be wary of backing them ‘blind’ in the future. My intention is simply to give more credence to the bias rather than encouraging you to be backing them in this way. Obviously if you fancy a horse or even two horses that are well drawn then that is different – my advice then would be to back them. Good luck!

 

 

 

David Renham  

 

Do you miss the original Part 1 of this article?

Read it here = > THE DRAW IN UK FLAT RACING – A BROAD OVERVIEW

 

----------------------------------------------------------------------

This article was written by Dave but first appeared on the interesting OnCourseProfits.com website / magazine.