Posts

Dave Renham: A synopsis of 6f and 7f AW Draw/Pace

In my last article I examined draw and running style combinations in five-furlong handicaps on the all-weather, with the main focus on front runners (those horses that take the early lead).

That article showed that on turning AW courses over the minimum trip (8+ runners), it was much easier to lead early from a lower draw compared to a higher one. That much is simple geometry: horses drawn low are closest to the rail and hence have less distance to travel to the first corner than their wider-drawn counterparts. It is worth noting that the positioning of the first bend can make a difference, as can the tightness of the turn.

However, the most surprising finding from the first article was that higher drawn horses that take the early lead actually go on to win more often than early leaders drawn low. I still cannot quite get my head round why this may be the case. As stated in that previous piece, I have always assumed that it is likely to have been quite an effort to pass so many horses to get to the lead from a wide draw. In addition to this, these runners probably would have had to run slightly further to achieve this.

Since writing the article I have tried to come up with a logical explanation for why higher-drawn horses have been able to win more often when leading early. Perhaps once these wide drawn runners get to the lead, the jockey on board tries to slow the pace down slightly in order to give his horse a breather, knowing that it would expended more energy than is ideal over that first half furlong or so.

More likely, though, is the impact of physics. As can be seen from the crude mock up below, a horse drawn inside has the best chance to get to the turn in front because it has the least distance to travel; but, once it gets to the turn the horse may need to decelerate in order to navigate around. Conversely, although a wider drawn runner has less chance to reach the turn in front - due to the potential of other horses inside to show early speed - on the occasions that a wide-drawn horse faces no pace contention, that horse can negotiate more of the turn at greater speed due to the angle at which it approaches the bend.

This of course depends on the location of the bend in relation to the start of the race. There is also a rule about jockeys staying in lanes for 100 yards, which might be described as 'loosely observed'. Regardless, hopefully it is clear how the less frequent wide drawn leader might win more often.

The impact of stall position on speed into the first turn

The impact of stall position on speed into the first turn

 

This is simply conjecture but in certain cases this could be what is happening. It might another day be worth looking at the new sectional timing data on Geegeez and matching it to those races where wide drawn runners had led early and gone onto win.

In this article I am going to look at six- and seven-furlong handicaps to see if similar patterns emerge in terms of draw / front runner combinations. Newcastle will be ignored as these distances are raced on a straight track there, but I will include Southwell this time as the six and seven furlong trips are raced around a bend there. Thus, we have six courses to look at: Chelmsford, Dundalk, Kempton, Lingfield, Southwell and Wolverhampton.

All-weather 6f handicaps (8 + runners)

Let's start by looking at draw / front runner combinations over six furlongs in handicaps. I only ever use handicap races for this type of research as non-handicap data is far less reliable. As mentioned in the first piece, the draw is split equally in three – low, middle and high - and hence one would expect, given a level playing field, that the ‘led early’ percentages would hit around 33.3% respectively from each section.

wdt_ID Course & Distance Low drawn led% Middle draw led% High draw led%
1 Chelmsford 6f 40.23 33.20 26.56
2 Dundalk 6f 53.13 28.13 18.75
3 Kempton 6f 40.32 32.66 27.02
4 Lingfield 6f 41.16 32.43 26.40
5 Southwell 6f 42.30 36.39 21.31
6 Wolverhampton 6f 34.53 30.55 34.92

For five of the six courses we see that once again the early leader is more likely to come from the lowest third of the draw – those drawn closest to the inside rail. Only Wolverhampton bucks the trend and this is probably because the first bend is more than a quarter mile from the start. That presents less of a positional advantage to the inside stalls and, essentially, the quickest horse from the gates should lead regardless of draw position. Dundalk seems to favour lower drawn horses the most with the bottom third of the draw producing more than half of all early leaders under these conditions.

The following table is another way of illustrating how much more likely low drawn horses are to lead than high drawn ones – I used this approach in the previous article and have replicated it for this range. It has been calculated by dividing the "low draw led%" by the "high draw led%".

wdt_ID Course & Distance Low% / High%
1 Chelmsford 6f 1.51
2 Dundalk 6f 2.83
3 Kempton 6f 1.49
4 Lingfield 6f 1.56
5 Southwell 6f 1.98
6 Wolverhampton 6f 0.99

Compared to the five furlong data these figures are not as high, but nevertheless if you are keen to predict the front runner, which we know is potentially a profitable angle, then horses from lower stalls do lead early significantly more often than higher drawn ones.

If we take Wolverhampton out of the equation and focus on the other five courses at six furlongs, when we increase to 12 or more runners the front running bias to lower draws does increase:

wdt_ID Low drawn led% Middle draw led% High draw led%
1 45.80 34 20.20

Under these circumstances the lowest third of draw is around 2.3 times more likely to produce the early leader of the race. This stronger bias mirrors the data we saw when analysing five-furlong handicaps. With higher draws starting further away from the inside rail in bigger fields, it is even harder for such horses to get to the early lead.

For the record here are the figures for Wolverhampton, where there have been over 300 qualifying races:

wdt_ID Low drawn led% Middle draw led% High draw led%
1 33.79 34.47 31.74

A very even split – with that long run to the first bend it seems that bigger fields do not make it more difficult for high drawn horses to lead early.

Moving on, let us now look at win percentages for the early leaders from each third of the draw at the six courses. Here is the six-furlong handicap data for eight or more runners:

wdt_ID Course & Distance Low drawn leaders race win% Middle draw leaders race win% High draw leaders race win%
1 Chelmsford 6f 20.39 17.65 20.59
2 Dundalk 6f 17.65 12.96 16.67
3 Kempton 6f 13.50 16.67 15.67
4 Lingfield 6f 16.67 17.95 25.20
5 Southwell 6f 18.60 18.92 23.08
6 Wolverhampton 6f 10.39 13.97 12.78

I mentioned at the beginning of this article that in five-furlong handicaps (at Chelmsford, Dundalk, Kempton, Lingfield and Wolverhampton), wider/higher drawn horses that take the early lead rather surprisingly go onto win more often than horses leading early from low draws. Over this extra furlong we can see that the courses give us a more even profile in terms of eventual win percentage. Having said that higher drawn horses that lead early still win on average slightly more often than lower drawn leaders. At Lingfield and Southwell, for example, higher drawn horses that take the early lead go onto win roughly one race in every four.

This even looking playing field is replicated when we combine all the 12+ runner data. Merging all courses together we get these win percentages:

wdt_ID Low drawn leaders race win% Middle draw leaders race win% High draw leaders race win%
1 13.26 14.64 12.92

As I have already alluded to, before researching and writing the five-furlong article I had expected that five- to seven-furlong races run around a bend would give horses that led early from a low draw much more chance of winning than those from a high draw. I had expected this bias against higher drawn horses to get even stronger the further the horses had to travel. It seems that for this theory I was right at least – it is harder over six furlongs than five furlongs for higher drawn early leaders to win. However, I had not expected higher drawn horses to still be more successful, in 8+ runner races at least, than lower drawn ones.

 

All weather 7f handicaps (8 + runners)

Now let's look at draw / front runner combinations in seven-furlong all-weather handicaps:

wdt_ID Course & Distance Low drawn led% Middle draw led% High draw led%
1 Chelmsford 7f 40.97 26.87 32.16
2 Dundalk 7f 55.43 25.72 18.84
3 Kempton 7f 42.18 34.73 23.09
4 Lingfield 7f 41.18 28.88 29.95
5 Southwell 7f 48.01 31.13 20.86
6 Wolverhampton 7f 49.74 27.37 22.88

All six courses this time show that the early leader is more likely to come from the lowest third of the draw. Dundalk once again provides the strongest bias, while at Southwell and Wolverhampton it is significant too. To perhaps illustrate this more clearly I have once again created a table showing the figure that is calculated by taking the low draw led% and dividing it by the high draw led%:

wdt_ID Course & Distance Low% / High%
1 Chelmsford 7f 1.27
2 Dundalk 7f 2.94
3 Kempton 7f 1.83
4 Lingfield 7f 1.38
5 Southwell 7f 2.30
6 Wolverhampton 7f 2.17

Interestingly it seems that in general lower drawn horses find it easier to lead over seven furlongs than at six. Again, on some tracks, notably Wolverhampton where the seven furlong start is in a chute on the brow of a bend, geometry plays its part.

Increasing field size to twelve or more runners enhances the front-running bias of lower drawn horses (as it does, too, over five- and six-furlongs).

wdt_ID Low drawn led% Middle draw led% High draw led%
1 49.31 29.29 21.40

Once again bigger fields give lower drawn horses a better chance of leading early.

The below shows the win percentage of early leaders from each third of the draw at the six courses over seven furlongs (8+ runner handicaps).

wdt_ID Course & Distance Low drawn leaders race win% Middle draw leaders race win% High draw leaders race win%
1 Chelmsford 7f 19.35 16.39 10.96
2 Dundalk 7f 12.42 12.68 11.54
3 Kempton 7f 14.48 10.44 10.74
4 Lingfield 7f 12.55 17.90 17.26
5 Southwell 7f 15.86 18.09 14.29
6 Wolverhampton 7f 11.50 13.81 11.16

This is quite an even set of figures when looking at the courses as a whole. Nevertheless, we still see that higher drawn horses which lead early are not at any real disadvantage. Those general themes are still true when we combine all the 12+ runner data from seven-furlong all-weather handicaps. Grouping the six turning courses we get these win percentages:

wdt_ID Low drawn leaders race win% Middle draw leaders race win% High draw leaders race win%
1 13.22 11.60 13.30

 

Summary

The above contains some interesting insights which may be combined with what we learned about five furlong handicaps on the all weather last time.

Like with races at the minimum, it may be easier to get to the lead from a lower draw over six and seven furlongs, but don’t be put off by a potential front runner drawn high. If your wide-drawn horse does lead, it has just as much chance of going onto win as a front runner drawn low.

Although these are not quite the startling statistics from the five-furlong article, to my eyes some of the findings are still surprising.

Personally, I am still shaking my head not quite believing what I have discovered over the past two articles. I just wonder how many bets I have ignored over the years due to a potential front runner being drawn high. Far too many!

- DR

Dave Renham: A Synopsis of 5f AW Draw/Pace

I have discussed pace angles in numerous Geegeez articles – see this list – and once again I would like to revisit this key area, this time in conjunction with draw, writes Dave Renham.

I have noted before that if you were able to predict the front runner in certain types of races it would amount to a license to print money. For example, going back to 2011, if you managed to correctly predict the front runner in every all-weather UK 5f handicap race with 8 or more runners, you would have profited by over 60p for every £1 staked!

Indeed at Kempton Park the profit would have been £1.04 for every £1 staked. For the record, in 6f handicaps on the sand you would have also profited from front runners to the tune of 33p for every £1 staked, while in 7f handicaps you still would have made 17p per £1 staked.

Naturally, and unfortunately, predicting who will lead in all-weather sprint handicaps is not as easy as all that.

In the past I have looked at different ideas to help increase the chances of predicting the front runner. For example, looking for horses that had led LTO, or looking for horses that have the highest pace score average over the past four races. I have also studied going conditions, the effect of field size etc.

One area though that I have yet to look at in real depth is the position of horses in terms of the draw. For this piece I have collated some all-weather handicap stats from the draw analyser on Geegeez, which also contains draw / run style data.

The draw can have a significant effect at some courses in both a positive and negative way. Races where the first bend is close to the start should offer lower drawn horses some advantage as they are berthed closest to the inside. At the tight turning course of Chester for example, this low draw bias is well known and documented.

Just as there can be a potential draw bias due to being drawn closest to the inside rail, one would assume that these horses have a greater chance of leading early. This is simply due to the fact that they have less distance to travel to the rail at the first corner than horses drawn wider. Of course, not all horses will try to lead early, but I felt it was time to crunch the numbers as I believed the data would back up my theory.

For the record, I have included Irish course Dundalk along with the six UK all weather tracks.

All weather 5f handicaps (8 + runners)

Let us begin by looking at draw / run style combinations over 5f. The draw is split equally in three – low, middle and high - and hence one would expect, given a level playing field, that the ‘led early’ percentages would hit around 33.3% respectively from each section.

It should also be noted that 5 of the 7 course and distances are run round a bend with only Newcastle and Southwell run on a straight course. A look at Newcastle and Southwell first:

wdt_ID Course & Distance Low drawn led% Middle draw led% High draw led%
1 Newcastle aw 5f 41.67 39.81 18.52
2 Southwell aw 5f 29.55 38.64 31.82

The Southwell figures are relatively even which is what I would have expected. However, the Newcastle stats are interesting with higher drawn horses far less likely to lead than those drawn low to middle. I cannot give a reason why this is the case, but it will be interesting to see if this pattern continues in the coming years.

Onto the other five courses and for the remainder of this article I will just focus on these as all distances are on a turning strip:

 

wdt_ID Course & Distance Low drawn led% Middle draw led% High draw led%
1 Chelmsford 5f 43.65 37.30 19.05
2 Dundalk 5f 45.74 37.98 16.28
3 Kempton 5f 41.45 35.53 23.03
4 Lingfield 5f 46.12 33.47 20.41
5 Wolverhampton 5f 44.14 35.67 20.20

 

This table shows that at all five courses the early leader is more likely to come from the lowest third of the draw – those drawn closest to the inside rail. I am pleased the stats seem to back up my original theory. In addition, horses from the middle stalls lead more often than those drawn high, suggesting there is a correlation between draw position and likelihood of leading.

The following table gives another way of illustrating how much more likely low drawn horses are to lead than high drawn ones – this has been very simply calculated by dividing the low draw led% by the high draw led%:

wdt_ID Course & Distance Low% / High%
1 Chelmsford 5f 2.29
2 Dundalk 5f 2.81
3 Kempton 5f 1.80
4 Lingfield 5f 2.26
5 Wolverhampton 5f 2.19

This table illustrates the bias to lower drawn front runners quite neatly with four of the five featured tracks’ minimum distance handicaps seeing lower drawn horses more than twice as likely to lead early as higher drawn ones. Dundalk seems to have the strongest low drawn front running bias and it is also worth sharing that horses drawn 1 and 2 at the Irish venue have provided the early leader 31% of the time.

Combining the data for all round-course 5f handicaps on the all-weather, and increasing the field size to 12 or more runners, there is an even stronger bias to low draws leading early. There are over 170 qualifying races which is a decent enough sample:

 

wdt_ID Low drawn led% Middle draw led% High draw led%
1 49.13 36.99 13.87

 

Under these circumstances the lowest third of draw are around 3.5 times more likely to produce the early leader of the race. This stronger bias makes sense as higher draws start even further away from the inside rail in bigger fields.

Another assumption I wanted to validate was that when higher drawn horses lead early they are less likely to go onto win: the reasoning behind this is that I perceived it to have generally been quite an effort to pass so many horses to get to the lead from a wide draw, as well as the fact that such runners would probably have had to travel slightly further to achieve this. Combining these factors, it would be logical to deduce that the horse might tire late on due to its earlier exertions in getting to the lead.

However, the stats do not back this up. Below are the win percentages for early leaders from each third of the draw at the five round-course all-weather tracks, firstly focusing on 8+ runner handicap data:

 

wdt_ID Course & Distance Low drawn leaders race win% Middle draw leaders race win% High draw leaders race win%
1 Chelmsford 5f 18.18 25.53 37.50
2 Dundalk 5f 15.25 12.24 28.57
3 Kempton 5f 28.57 22.22 28.57
4 Lingfield 5f 24.78 24.39 16.00
5 Wolverhampton 5f 18.08 15.53 18.55

 

Horses that lead from high draws at Chelmsford manage to go on to win three races in eight; those at Dundalk and Kempton prevail better than one in four. Only at Lingfield does it seem a negative to lead early from a high draw.

A similar pattern emerges when we look at the 12+ runner handicap data. Combining the courses we get these win percentages:

 

wdt_ID Low drawn leaders race win% Middle draw leaders race win% High draw leaders race win%
1 12.94 12.50 20.83

 

I concede these stats have really surprised me. However, in many respects this is good news if you like backing front runners. In the past I may have been put off by a potential front runner drawn wide as I would have assumed if they did manage to lead they were less likely to win. This is not the case –over 5 furlongs at these courses anyway!

Conclusions

This article has shown that in all-weather 5f handicaps contested on a round course, it is easier to lead from a lower draw than a higher one, BUT… in terms of winning the race you may prefer your potential front runner to be drawn high!

Food for thought I hope, and if you have enjoyed this piece you will perhaps be pleased to know that I plan to look at 6f handicaps in a follow-up article.

  • DR

p.s. if you want to understand the impact of draw and pace in combination, Geegeez Gold's new Heat Map underlay within the pace tab does just that, for the specific course/distance/field size/race type combination in question - example below. Click here to join Geegeez Gold >