Bookmakers place odds are worse than I thought.

The idea behind this blog was to conduct in-depth of study of place betting on horses, and discover whether placed odds offered any betting value. 

For the purpose of this blog, it will be assumed in all cases that odds accurately reflect each horse's chance of winning. 

The same odds will then be used to estimate the place odds as accurately as possible, you can then decide whether to amend your betting habits in relation to place betting

Books will be adjusted to 100% and decimal odds used in all cases to make easy and fair comparisons with the true odds.

I shall use an extreme case as an example to show what I'm trying to achieve. 

At this point, I have no preconceived idea as to what the conclusions may show, answers will, therefore, be totally unbiased.

The most basic of basics

This is the most hypothetical 8 runner race ever created and would be classed by punters as a good each-way race, and by the bookmakers as a bad each-way race.

Horse	Betting Odds	Percentage Chance
Frankel	1/1 OR 2.0	50%
Mill Reef	6/4 OR 2.5	40%
Brigadier Gerard	9/1 0R 10.0	10%
Pinatubo	1000	0.10
Swain	1000	0.10
Sea The Stars	1000	0.10
Galileo	1000	0.10
Rock Of Gibraltar	1000	0.10

Five of the eight runners are BSP 1000, with the odds for the other three being Evens, 6/4 and 9/1 which in decimal becomes 2.0, 2.5 and 10.0.

The even-money chance is not 9 times as likely to win as the 9/1 shot. 

Neither is the 6/4 chance 6 times as likely. 

Using decimal odds it is far simpler to calculate their true chance of winning, simply divide 100 by the odds. 

In this case ( 2, 2.5 and 10 ) those percentages are 50, 40 and 10. 

Now it becomes clear that the even-money chance is only 5 times as likely to win as the 9/1 chance. 

The 6/4 chance is 4 times as likely.

Barring accidents these three would finish 1st, 2nd and 3rd in some order so the 1/5 odds that you would ( or in this instance wouldn't ) get from the bookies bear no resemblance to the true odds. 

A truly extreme example of what is deemed a “bad each-way race” by the bookmakers. 

So how does this relate to the true place odds?

The next step is to take races of various numbers of runners and differing odds ranges to see what happens. 

As I have only a vague idea of what to expect, the samples will be genuinely random other than using figures easy to handle wherever possible. 

In each instance, I shall use a tight market and a loose one.

For an easy start, I shall take a 5 runner race with ¼ odds for 2 places.

The formula used may not be exact scientifically and almost certainly not an exact representation of how a race may turn out given sufficient trials but correct arithmetically and with a degree of logic. 

It is easy to calculate the arithmetical winning chance of horses at various odds so that was the obvious starting point.

Then taking the favourite, Andy, as winning 33.33% of races (actually 33.16% when rounding to a 100% book ) 

I calculated the chance of each horse finishing second to Andy in a ratio of their win odds, hence not 100% accurate other than arithmetically.

This was then done with the other 4 horses winning in turn. I was delighted when the figures added up to 200 in this case and between 199.98 and 200 in all others.

The 5 runners and odds for the tight race are

• Frankel 3.00
• Mill Reef 4.50
• Brigadier Gerard 5.50
• Pinatubo 7.00
• Swain 8.00

Without boring you with the actual data, the overall place odds are given to a 100% win book ( which doesn't happen ) so a 200% place book and ¼ odds with ‘arithmetical' odds in brackets.

• Frankel 1.50 ( 1.66 )
• Mill Reef 1.88 ( 2.23 )
• Brigadier Gerard 2.13 ( 2.65 )
• Pinatubo 2.50 ( 3.28 )
• Swain 2.75 ( 3.71 )

The place book is arithmetically 200% yet bookies odds amount to a rip off at 243.42%, and that is without any SP over round built-in. 

When I worked in the trade, these races were 1/3 odds 2 places for 6/7 runners and from memory win only for 5 runners. 

The 5 runners and odds for the loose race are

• Frankel 1.80
• Mill Reef 5.00
• Brigadier Gerard 8.00
• Pinatubo 12.00
• Swain 25.00

The same calculations produce the following.

• Frankel 1.20 ( 1.19 )
• Mill Reef 2.00 ( 2.00 )
• Brigadier Gerard 2.75 ( 3.07 ) 
• Pinatubo 3.75 ( 4.51 )
• Swain 7.00 ( 9.23 )

The place book is arithmetically 200% and bookies odds amount to an almost acceptable over round of 210.65%, but again that is without any SP over round built-in. 

That was fun so now for 6 runners

Sea The Stars has joined the races and everything was calculated as per the 5 runner races. The 6 runners and odds for the tight race are

• Frankel 3.75
• Mill Reef 4.50
• Brigadier Gerard 6.00
• Pinatubo 7.50
• Swain 9.00
• Sea The Stars 10.00

As anticipated, the extra runner did little other than create greater over rounds for the bookies on the basis of still using a 100% book, but having one more horse on their side ( 4 losers rather than 3 ).

The bookie's odds and arithmetical odds are portrayed as in the above examples.

• Frankel 1.69 ( 2.00 )
• Mill Reef 1.88 ( 2.30 )
• Brigadier Gerard 2.25 ( 2.94 )
• Pinatubo 2.63 ( 3.60 )
• Swain 3.00 ( 4.26 )
• Sea The Stars 3.25 ( 4.70 )

That is still an arithmetical 200% place book but the bookies now have an over-round of 259.23%. So what happens when we put in an odds on favourite again? The runners with odds are now

• Frankel 1.91
• Mill Reef 5.00
• Brigadier Gerard 8.00
• Pinatubo 13.00
• Swain 21.00
• Sea The Stars 34.00

Calculating place odds on the established formula produced the following figures.

• Frankel 1.23 ( 1.23 )
• Mill Reef 2.00 ( 2.09 )
• Brigadier Gerard 2.75 ( 3.20 )
• Pinatubo 4.00 ( 5.08 )
• Swain 6.00 ( 8.10 )
• Sea The Stars 9.25 ( 13.03 )

Yes, a reduction on the tight race but still very high at 220.31% excluding the SP over round. 

I have purposely not looked at any actual races similar to these so as not to be influenced in any way. I shall check some when I have time and then proceed to do the same exercise with bigger fields and a much wider spreadsheet. 

And then there were 7 runners

Galileo has just turned up and offered to make up the numbers.

There is no need to repeat the format so I shall just show the figures.

The 7 runners and odds for the tight race are

• Frankel 4.00
• Mill Reef 5.00
• Brigadier Gerard 6.00
• Pinatubo 7.00
• Swain 10.00
• Sea The Stars 13.00
• Galileo  15.00

The extra runner seems to have had very little impact on either set of figures.

• Frankel 1.75 ( 2.14 )
• Mill Reef 2.00 ( 2.56 )
• Brigadier Gerard 2.25 ( 2.99 )
• Pinatubo 2.50 ( 3.43 )
• Swain 3.25 ( 4.78 )
• Sea The Stars 4.00 ( 6.13 )
• Galileo 4.50 ( 7.03 )

That is still an arithmetical 200% place book but the bookies now have an over-round of 269.58%.

So what happens when we put in an odds on favourite again? 

The runners with odds are now

• Frankel 1.91
• Mill Reef 5.00
• Brigadier Gerard 8.00
• Pinatubo 15.00
• Swain 21.00
• Sea The Stars 41.00
• Galileo 67.00

Calculating place odds on the established formula produced the following figures.

• Frankel 1.23 ( 1.23 )
• Mill Reef 2.00 ( 2.09 )
• Brigadier Gerard 2.75 ( 3.20 )
• Pinatubo 4.50 ( 5.84 )
• Swain 6.00 ( 8.10 )
• Sea The Stars 11.00 ( 16.57 )
• Galileo 17.50 ( 25.51 )

That is still an arithmetical 200% place book but the bookies now have an over-round of 221.52%.

You can't please everyone all the time even with 8 runners

When Rock Of Gibraltar appeared from nowhere, the bookmakers were noticeably less happy than the punters. 

ROG was shall we say, rather poor, and was always the outsider of the field. 

However, his presence meant the bookies' odds would not change much but they would now have to pay three places to the each-way punters in an 8 runner race. 

Needless to say, this had the potential to turn the odds against them in certain circumstances.

While I suspect my arithmetical model is less accurate with 3 places ( until I can be sure it isn't ) the general pattern seems to remain.

The format is the same but now 1/5 odds 3 places.

Runners and win odds for a tight race

• Frankel 3.75
• Mill Reef 5.50
• Brigadier Gerard 6.50
• Pinatubo 8.00
• Swain 10.00
• Sea The Stars 13.00
• Galileo 17.00
• Rock Of Gibraltar 26.00

Place odds

• Frankel 1.55 ( 1.49 )
• Mill Reef1.90 ( 1.89 )
• Brigadier Gerard 2.10 ( 2.13 )
• Pinatubo 2.40 ( 2.51 )
• Swain 2.80 ( 3.04 )
• Sea The Stars 3.40 ( 3.83 )
• Galileo 4.20 ( 4.90 )
• Rock Of Gibraltar 6.00 ( 7.33 )

With 3 places the arithmetical place book changes to 300% and the bookies' over-round are 312.04%, but of course with no SP overs taken into account

Runners and win odds for the loose race

• Frankel 1.91
• Mill Reef5.50
• Brigadier Gerard 9.00
• Pinutabo 14.00
• Swain 21.00
• Sea The Stars 33.00
• Galileo 51.00
• Rock Of Gibraltar 67.00

Place odds

• Frankel 1.18 ( 1.07 )
• Mill Reef1.90 ( 1.48 )
• Brigadier Gerard 2.60 ( 2.04 )
• Pinutabo 3.60 ( 2.96 )
• Swain 5.00 ( 4.29 )
• Sea The Stars 7.40 ( 6.60 )
• Galileo 11.00 ( 10.07 )
• Rock Of Gibraltar 14.20 ( 13.16 )

With 3 places the arithmetical place book is 300% and the bookies' over-round is 253.12%, but of course with no SP overs taken into account. 

Even so, that is a massive swing in favour of the punter. 

To combat this, bookies will often amend their place terms or restrict stakes that may be placed. 

Once again though, it does show the better value to be had with the shorter priced horses, to read more about this subject check out our article on what is betting value, which is packed with helpful info. 

The fact that this is consistent will in part be due to it being done to a formula.

Having got a background feeling for what to expect. 

I can vouch for the accuracy of the arithmetic however if you spot a flaw please let me know

I shall now set about Saturday's racing ( Aug 22nd ) using SP and BSP for comparison. 

I was away from home all day yesterday so findings will not be biased by any prior knowledge. 

Results for that will appear on the "Each Way" thread.