Draw Biases

Horse racing chat. Betting issues, racing news

Draw Biases

Postby loser2winner » Tue Mar 20, 2012 4:45 pm

Imo people on the forum say a horse has poor draw

I'm not sure what there looking at

IMO I look at the stalls either side of the draw if they are red then you could say poor draw but if there not red either side then IMO it don't mean a poor draw even if that stall is red

And if it's not a hcp does the draw mean as much than a hcp draw I don't think so imo
loser2winner
 

Re: Draw Biases

Postby Pittsburgh Phil » Tue Mar 20, 2012 5:35 pm

Depends on what is meant by a poor draw.

Most punters / pundits say a horse is poorly drawn if it is running from a part of the course that rarely wins (high at Chester). That's fair enough but it doesn't mean 'horse can never win from that draw'.

This site uses value stats to determine which draw range is value. Red and those stalls tend to provide poor value. Blue and they tend to provide good value. This is totally different from can / can't win from stalls XYZ.

Stall Ranges
Never look at one stall and state that it is good or bad. Whilst there may be some bogey stalls on the extreme left or right it is total mumbo jumbo to say that Stall 5 is lucky / unlucky. Leave that nonsense to Big Mac.

It is far better to look at a range of stalls and determine if that group is positive or negative. This site splits the draw ranges into thirds so that there are stats for low, mid high.

Match Races
Draw biases will vary depending on the going and field size so when analysing you should try to match that. But never be too exact else you will get a much smaller sample and may not actually be as accurate as you think.

If the going is described as Good to Soft then analyse Soft, Good to Soft and Good as one group. By including the ranges either side of the 'official going' you can smooth out any errors in that 'official going'.

Don't analyse exactly 14 runners in a 14 runner race. It's best to use ranges such as 8-10, 11 to 15, 16-20, 21+. For simplicity this site only recognises the following field sizes:

0-9 No Draw bias determined (it's too low to analyse)
10-15 Medium field size
16+ Big field size

It's best to only include handicaps when analysing draw biases as these give closer finishes and are thus more useful for identifying draw biases. Only include stakes if the sample size is not sufficient. This doesn't mean you should not use draw biases in stakes - the draw bias will still be there. It means that only analysing draw biases in handicaps will make them stick out more.

Weed out outsiders. They do nothing for draw bias analysis. Ideally you should not include any horses that finished 30+ lengths behind the winner.

Draw Biases are Permanent, Golden Highways are Temporary
Temporary draw biases can exist but they are always man made. Be aware of which courses they can happen at and always remember it. Remember too why it happened. At

Newmarket one year draw biases were temporary because the local trainers wanted the grass mown a certain way. Another year temporary biases existed on the July course because they changed the way the service vehicles go down to the start.
User avatar
Pittsburgh Phil
Chief
 
Posts: 16955
Joined: Jan 2003
Location: UK

Re: Draw Biases

Postby Pittsburgh Phil » Tue Mar 20, 2012 5:46 pm

Flatstats Draw Bias Resources
The racecards and daily stats page show the Draw Bias Indexes where sufficient data is available (field size, going etc.)

Draw Bias on Systems Sheet / Racecards

Today at Kempton this shows:

3:40K Low +15, Mid +18, High -3
5:40K Low +19, Mid +7, High -18


As I said. This doesn't mean 'horse can't win from high'. It means that those drawn high do not win as their odds would expect them to. They are overbet / tend to be poor value bets.

Draw Stats Reports
This is a form, which you can set up for any course and distance and filter for extra parameters such as field size, type of race, price etc. Information is shown in chart form and tables and is very easy to use.

Flatstats Draw Stats

Draw Bias Tables
Want to compile a list of courses, distances and goings that have the strongest draw biases? Use this table!

Rather than you clicking on each course, going, distance, and field size this table has already done it for you. Currently over 500 different combinations are available to view and they are sortable by A/E or profit.

As of today the course / distance / going / field size combinations with the most value draw bias range on the turf are:

Doncaster - Good - 6f - Medium Field Size - Low = 1.53 A/E
Salisbury - Firmish - 8f - Medium Field Size - Mid = 1.47 A/E
Chepstow - Good - 8f - Medium Field Size - Low = 1.39 A/E


Not a Chester in sight! The first Chester record is 18th on the list. Now how many punters would think that Chester has the strongest draw bias?

It may produce more winners than other courses / distances but they are not as valuable as those three listed above.

Go to page 24 of that table (or sort by A/E descending) and you will see the worst draw bias ever.

Doncaster - Good - 6f - Large Field Size - Mid = 0.12 A/E

That is an extremely poor A/E stat. You are not going to get a value run for your money if your horse is drawn there in that particular type of race.

Have a look at the tables. Setup alerts for the best and worst. Download the data to CSV if you wish.

Course Draw Bias Tables
User avatar
Pittsburgh Phil
Chief
 
Posts: 16955
Joined: Jan 2003
Location: UK

Re: Draw Biases

Postby loser2winner » Tue Mar 20, 2012 6:43 pm

PP
Thanks
loser2winner
 

Re: Draw Biases

Postby Eyecatcher » Thu Mar 22, 2012 7:49 am

Thankyou very much for that.. excellent post @Pittsburgh Phil .. .. one of the best for a while..

by the way, what is the period about "stats figures"??

Thanks in adance.

Cheers

Best Regards.
User avatar
Eyecatcher
7lb Claimer
 
Posts: 12
Joined: Apr 2011

Re: Draw Biases

Postby SirKnight » Thu Mar 22, 2012 11:12 am

Hi Phil,

Thanks for a great post.

I have one question do you think the Newmarket draw bias stats could be skewed by the July course or vice versa?

Regards

SirK
User avatar
SirKnight
5lb Claimer
 
Posts: 230
Joined: Aug 2003

Re: Draw Biases

Postby Pittsburgh Phil » Thu Mar 22, 2012 11:21 am

period about "stats figures"?


Do you mean the timescale? All stats on this site are based on a rolling 10 years so that you can see stats for the past 10 years up until yesterday.
User avatar
Pittsburgh Phil
Chief
 
Posts: 16955
Joined: Jan 2003
Location: UK

Re: Draw Biases

Postby Pittsburgh Phil » Thu Mar 22, 2012 11:23 am

SirK the draw biases at Newmarket are only apparent at the longer distances where the Rowley and July courses share the same part of the track.

Newmarket July and Rowley Mile Courses
User avatar
Pittsburgh Phil
Chief
 
Posts: 16955
Joined: Jan 2003
Location: UK

Re: Draw Biases

Postby Eyecatcher » Thu Mar 22, 2012 11:39 am

Pittsburgh Phil wrote:
period about "stats figures"?


Do you mean the timescale? All stats on this site are based on a rolling 10 years so that you can see stats for the past 10 years up until yesterday.


Yes, about timescale.. Thanks again Sir.
User avatar
Eyecatcher
7lb Claimer
 
Posts: 12
Joined: Apr 2011

Re: Draw Biases

Postby Pittsburgh Phil » Thu Mar 22, 2012 11:50 am

Yes the full 10 years are available.

You can of course filter the results for any date range you wish: The last 5 years, last 2 years, the last 3 months.

This is useful for checking temporary draw biases which can be apparent such as on the all weather during long periods of frost.
User avatar
Pittsburgh Phil
Chief
 
Posts: 16955
Joined: Jan 2003
Location: UK

Re: Draw Biases

Postby pippex » Thu Mar 22, 2012 2:27 pm

In my opinion draw advantage is linked to pace,a bad draw may actually be an advantage to a hold up horse.Some courses favour hold up horses e.g. doncaster.In large fields the position of the front runners on the track,may appear to produce a draw bias where as it was a pace bias. Close study of the last 4 all aged handicaps over the course and distance can sometimes give profitable clues.At some courses horses who get to the rail first win..e.g Newcastle.
User avatar
pippex
Most Helpful Poster 2012
 
Posts: 21483
Joined: Jan 2003

Re: Draw Biases

Postby wim » Sat Mar 24, 2012 2:25 pm

I noticed something today that happens when there are 11 runners. Draw 1 is given loads of space because the stalls are in sets of 10. So I thought I would check that stats at Lingfield.

6f D1:
10 runners Win 7.9%
11 runners Win 9.9%
12 runners Win 6.9%

7f D1:
10 runners Win 5.5%
11 runners Win 11.7%
12 runners Win 5.2%

8f D1:
10 Runners Win 6.6%
11 runners Win 9.8%
12 runners Win 6.2%
User avatar
wim
Best Post and Best Tipster 2010
 
Posts: 5276
Joined: Mar 2007
Location: leicestershire

Re: Draw Biases

Postby jth » Sat Mar 24, 2012 2:34 pm

_goodpost:
Didnt help Jamie on Gatepost but v useful stat- thanks
User avatar
jth
Best Poster 2012
 
Posts: 19489
Joined: Jan 2003
Location: london

Re: Draw Biases

Postby pippex » Sat Mar 24, 2012 3:25 pm

excellent Wim _gc

just shows how valuable watching races can be
User avatar
pippex
Most Helpful Poster 2012
 
Posts: 21483
Joined: Jan 2003


Return to Horse Racing Chat

Who is online

Registered users: Aioros, chunneyball, Google Adsense [Bot], lombie
Big Priced Winner! Top Rated Dialogue Wins 11/1 | Subscriptions just £25 a month or see special offers  Subscribe Today!