The Best Staking Plan For Horse Racing

[Gingertipster]

What do you think the best staking plan is?
What percentage of bank do you risk on one horse or race?
What is your staking plan?

Mine is based on three things:
1) Probability of winning in percentage terms. More chance of winning the more money staked.
2) How much value I believe is in the bet. The more value the more money staked.
3) Size of my betting bank. The more money I’ve got the more money is staked

Well, four… The fact I am naturally cautious.

My Staking Plan:
My idea of the horse’s percentage chance of winning…
Minus the best odds available (in percentage terms)…
Times by (T)… Where (T) = Number of thousands of pounds in betting bank (to nearest thousand).
Plus (+)…
My idea of the percentage chance times by (C)… Where (C) = T number of thousands in bank divided by 5
Equals number of pounds staked (S)

This works out that: For any individual horse backed the stake can be as little as 0.1% of my betting bank on a rank outsider that I consider value… and up to a maximum of around 2.5% for an individual horse at a “shortish” price that I believe should be considerably shorter. However, if I think a horse is value I back it in some way, so usually have more than one bet in a race (main bet/s and savers,where main bets are what I consider to be best value in the race). Therefore can be up to around 5% of betting bank invested in any one race. Sometimes a little extra if believing more than one are exceptional value, but that’s extremely rare.

So a horse I believe has a 25% chance (fair 3/1) who is available 4/1 (20%) would have 1% of my betting bank staked… Shown below.

For a betting bank of £5,000 a horse I believe has a 25% chance of winning available at 4/1 (20%) is:
25% – 20% = 5
…and 5 being the number of thousands of pounds in the bank (T)
Therefore 5 X 5 = 25
Plus the result of:
5 thousands divided by 5 = 1
That 1 X my idea of the percentage chance of winning 25 = 25
So…
25 + 25 = 50
£50 staked @ 4/1 Returning £250 and £200 Profit.

For a betting bank of £7,000 the same 25% chance available at 4/1 (20%) is:
25% – 20% = 5
…and 7 being the number of thousands in the bank
Therefore 5 X 7 = 35
Plus the result of:
7 thousands divided by 5 = 1.4
That 1.4 X 25 = 35
So…
35 + 35 = 70
£70 staked @ 4/1 Returning £350 and £280 Profit

For a betting bank of £10,000 the same 25% chance available at 4/1 (20%) is:
25 – 20 = 5
10 being the number of thousands
5 X 10 = 50
Plus
10 ‘/, 5 = 2
2 X 25 = 50
50 + 50 = 100
£100 staked @ 4/1 Returning £500 and £400 profit.

For a betting bank of £15,000 the same 25% chance available at 4/1 (20%) is:
25 – 20 = 5
15 being the number of thousands
5 X 15 = 75
Plus
15 ‘/, 5 = 3
3 X 25 = 75
75 + 75 = 150
£150 staked @ 4/1 Returning £750 and £600 profit

For a betting bank of £20,000 the same 25% chance available at 4/1 is:
25 – 20 = 5
5 X 20 = 100
Plus
20 ‘/, 5 = 4
4 X 25 = 100
100 + 100 = 200
£200 staked @ 4/1 Returning £1000 and £800 profit

For a betting bank of £25,000 the same 25% chance available at 4/1 is:
25 – 20 = 5
5 X 25 = 125
Plus
25 ‘/, 5 = 5
5 X 25 = 125
125 + 125 = 250
£250 staked @ 4/1 Returns £1250 and £1000 Profit.

For a betting bank of £30,000 the same 25% chance available at 4/1 is:
25 – 20 = 5
5 X 30 = 150
Plus
30 ‘/, 5 = 6
6 X 25 = 150
150 + 150 = 300
£300 staked @ 4/1 Returns £1500 and £1200 Profit.

etc

Value Is Everything

[ricky lake]

Interesting. Good stuff

Rarely i back a horse to win nowadays

However the distances are my speciality

This year. I am convinced the best two jockeys around are called Doyle👍👍

Still the jumps are here now so its all guns blazing

My max bet is 10 pc of bank

My normal bet is about 2 pc

[Gingertipster]

Welcome back and good luck with the jumps season, Ricky.

Value Is Everything

[Gingertipster]

Here are a few other stakes from my staking plan:
A reminder of my strategy:

In my opinion the more chance a horse has, the more money should be staked.
The more value in a bet the more money should be staked…
However, also the fact less money is risked on outsiders should also be rewarded by greater profit.

For a betting bank of £10,000 a 50% chance available at 11/10 (47.6%) is:
50% – 47.6% = 2.4
2.4 X 10 = 24
2 X 50 = 100
24 + 100 = 124
£124 staked @ 11/10 Returns £260.40 and £136.40 Profit.

For a betting bank of £10,000 a 50% chance available at 5/4 (44.4%) is:
50 – 44.4 = 5.6
5.6 X 10 = 56
2 X 50 = 100
56 + 100 = 156
£156 staked @ 5/4 Returns £351 and £195 Profit.

For a betting bank of £10,000 a 50% chance available at 6/4 (40%) is:
50 – 40 = 10
10 X 10 = 100
2 X 50 = 100
100 + 100 = 200
£200 staked @ 6/4 Returns £500 and £300 Profit.

For a betting bank of £10,000 a 50% chance available at 7/4 (36.4%) is:
50 – 36.4 = 13.6
13.6 X 10 = 136
2 X 50 = 100
136 + 100 = 236
£236 staked @ 7/4 Returns £649 and £413 Profit.

………………………………………….
From the bets above you can see how the stakes and profit go up with value in the bet.
………………………………………….

For a betting bank of £10,000 a 40% chance available at 7/4 (36.4) is:
40 – 36.4 = 3.6
3.6 X 10 = 36
2 X 40 = 80
36 + 80 = 116
£116 staked @ 7/4 Returns £319 and £203 Profit.

For a betting bank of £10,000 a 40% chance available at 2/1 (33.3) is:
40 – 33.3 = 6.7
6.7 X 10 = 67
2 X 40 = 80
67 + 80 = 147
£147 staked @ 2/1 Returns £441 and £294 Profit.

For a betting bank of £10,000 a 40% chance available at 9/4 (30.8) is:
40 – 30.8 = 9.2
9.2 X 10 = 92
2 X 40 = 80
92 + 80 = 172
£172 staked @ 9/4 Returns £559 and £387 Profit.

………………………………………..

For a betting bank of £10,000 a 35% chance available at 85/40 (32%) is:
35 – 32 = 3
3 X 10 = 30
2 X 35 = 70
30 + 70 = 100
£100 staked @ 85/40 Returns £312.5 and £212.5 Profit.

For a betting bank of £10,000 a 35% chance available at 5/2 (28.6%) is:
35 – 28.6 = 6.4
6.4 X 10 = 64
2 X 35 = 70
64 + 70 = 134
£134 staked @ 5/2 Returns £469 and £335 Profit.

For a betting bank of £10,000 a 35% chance available at 3/1 (25%) is:
35 – 25 = 10
10 X 10 = 100
2 X 35 = 70
100 + 70 = 170
£170 staked @ 3/1 Returns £680 and £510 Profit.

………………………………………..

For a betting bank of £10,000 a 30% chance available at 11/4 (26.7%) is:
30 – 26.7 = 3.3
3.3 X 10 = 33
2 X 30 = 60
33 + 60 = 93
£93 staked @ 11/4 Returns £348.75 and £255.75 Profit.

For a betting bank of £10,000 a 30% chance available at 100/30 (23.1%) is
30 – 23.1 = 6.9
6.9 X 10 = 69
2 X 30 = 60
69 + 60 = 129
£129 staked @ 100/30 Returns £559 and £430 Profit.

For a betting bank of £10,000 a 30% chance available at 4/1 (20%) is:
30 – 20 = 10
10 X 10 = 100
2 X 30 = 60
100 + 60 = 160
£160 staked @ 4/1 Returns £800 and £640 Profit.

………………………………………..

For a betting bank of £10,000 a 25% chance available at 7/2 (22.2%) is:
25 – 22.2 = 2.8
2.8 X 10 = 28
2 X 25 = 50
28 + 50 = 78
£78 staked @ 7/2 Returns £351 and £273 Profit.

For a betting bank of £10,000 a 25% chance available at 9/2 (18.2%) is:
25 – 18.2 = 6.8
6.8 X 10 = 68
2 X 25 = 50
68 + 50 = 118
£118 staked @ 9/2 Returns £649 and £531 Profit.

For a betting bank of £10,000 a 25% chance available at 11/2 (15.4%) is:
25 – 15.4 = 9.6
9.6 X 10 = 96
2 X 25 = 50
96 + 50 = 146
£146 staked @ 11/2 Returns £949 and £803 Profit.

…………………………………………

For a betting bank of £10,000 a 20% chance available at 5/1 (16.7%) is:
20 – 16.7 = 3.3
3.3 X 10 = 33
2 X 20 = 40
33 + 40 = 73
£73 staked @ 5/1 Returns £438 and £365 Profit.

For a betting bank of £10,000 a 20% chance available at 6/1 (14.3%) is:
20 – 14.3 = 5.7
5.7 X 10 = 57
2 X 20 = 40
57 + 40 = 97
£97 staked @ 6/1 Returns £679 and £582 Profit.

For a betting bank of £10,000 a 20% chance available at 7/1 (12.5%) is:
20 – 12.5 = 7.5
7.5 X 10 = 75
2 X 20 = 40
75 + 40 = 115
£115 staked @ 7/1 Returns £920 and £805 Profit.

For a betting bank of £10,000 a 20% chance available at 8/1 (11.1%) is:
20 – 11.1 = 8.9
8.9 X 10 = 89
2 X 20 = 40
89 + 40 = 129
£129 staked @ 8/1 Returns £1161 and £1032 Profit.

………………………………………….

For a betting bank of £10,000 a 15% chance available at 13/2 (13.3%) is:
15 – 13.3 = 1.8
1.8 X 10 = 18
2 X 15 = 30
18 + 30 = 48
£48 staked @ 13/2 Returns £360 and £312 Profit.

For a betting bank of £10,000 a 15% chance available at 17/2 (10.5%) is:
15 – 10.5 = 4.5
4.5 X 10 = 45
2 X 15 = 30
45 + 30 = 75
£75 staked @ 17/2 Returns £712.5 and £637.5 Profit.

For a betting bank of £10,000 a 15% chance available at 10/1 (9.1%) is:
15 – 9.1 = 5.9
5.9 X 10 = 59
2 X 15 = 30
59 + 30 = 89
£89 staked @ 10/1 Returns £979 and £890 Profit.

For a betting bank of £10,000 a 15% chance available at 12/1 (7.7%) is:
15 – 7.7 = 7.3
7.3 X 10 = 73
2 X 15 = 30
73 + 30 = 103
£103 staked @ 12/1 Returns £1339 and £1236 Profit.

…………………………………………..

For a betting bank of £10,000 a 10% chance available at 11/1 (8.3%) is:
10 – 8.3 = 1.7
1.7 X 10 = 17
2 X 10 = 20
17 + 20 = 37
£37 staked @ 11/1 Returns £444 and £407 Profit.

For a betting bank of £10,000 a 10% chance available at 13/1 (7.1%) is:
10 – 7.1 = 2.9
2.9 X 10 = 29
2 X 10 = 20
29 + 20 = 49
£49 staked @ 13/1 Returns £686 and £637 Profit.

For a betting bank of £10,000 a 10% chance available at 15/1 (6.25%) is:
10 – 6.2 = 3.8
3.8 X 10 = 38
2 X 10 = 20
38 + 20 = 58
£58 staked @ 15/1 Returns £928 and £870 Profit.

For a betting bank of £10,000 a 10% chance available at 18/1 (5.3%) is:
10 – 5.3 = 5.7
5.7 X 10 = 57
2 X 10 = 20
57 + 20 = 77
£77 staked @ 18/1 Returns £1463 and £1386 Profit.

………………………………………….

For a betting bank of £10,000 a 5% chance available at 25/1 (3.8%) is:
5 – 3.8 = 1.2
1.2 X 10 = 12
2 X 5 = 10
12 + 10 = 22
£22 staked @ 25/1 Returns £572 and £550 profit.

For a betting bank of £10,000 a 5% chance available at 33/1 (2.9%) is:
5 – 2.9 = 2.1
2.1 X 10 = 21
2 X 5 = 10
21 + 10 = 31
£31 staked @ 33/1 Returns £1054 and £1023 Profit.

For a betting bank of £10,000 a 5% chance available at 50/1 (2%) is:
5 – 2 = 3
3 X 10 = 30
2 X 5 = 10
30 + 10 = 40
£40 staked @ 50/1 Returns £2040 and £2000 Profit.

………………………………………….

For a betting bank of £10,000 a 3% chance available at 50/1 (2%) is:
3 – 2 = 1
1 X 10 = 10
2 X 3 = 6
10 + 6 = 16
£16 staked @ 50/1 Returns £816 and £800 Profit.

For a betting bank of £10,000 a 3% chance available at 80/1 (1.2%) is:
3 – 1.2 = 1.8
1.8 X 10 = 18
2 X 3 = 6
18 + 6 = 24
£24 staked @ 80/1 Returns £1944 and £1920 Profit.

………………………………………….

For a betting bank of £10,000 a 2% chance available @ 80/1 (1.2%) is:
2 – 1.2 = 0.8
0.8 X 10 = 8
2 X 2 = 4
8 + 4 = 12
£12 staked @ 80/1 Returns £972 and £960 Profit.

For a betting bank of £10,000 a 2% chance available @ 125/1 (0.8%) is:
2 – 0.8 = 1.2
1.2 X 10 = 12
2 X 2 = 4
12 + 4 = 16
£16 staked @ 125/1 Returns £2016 and £2000 Profit.

…………………………………………..

For a betting bank of £10,000 a 1.5% chance available at 100/1 (1%) is
1.5 – 1 = 0.5
0.5 X 10 = 5
2 X 1.5 = 3
5 + 3 = 8
£8 staked @ 100/1 Returns £900 and £800 Profit

For a betting bank of £10,000 a 1.5% chance available at 125/1 (0.8%) is:
1.5 – 0.8 = 0.7
0.7 X 10 = 7
2 X 1.5 = 3
7 + 3 = 10
£10 staked @ 125/1 Returns £1260 and £1250 Profit.

…………………………………………..

For a betting bank of £10,000 a 1% chance available at 200/1 (0.5%) is:
1 – 0.5 = 0.5
0.5 X 10 = 5
2 X 1 = 2
5 + 2 = 7
£7 staked @ 200/1 Returns £1407 and £1400 Profit.

For a betting bank of £10,000 a 1% chance available at 500/1 (0.2%)
1 – 0.2 = 0.8
0.8 X 10 = 8
2 X 1 = 2
8 + 2 = 10
£10 staked @ 500/1 Returns £5010 and £5000 Profit.

Value Is Everything

[buckers]

Interesting stuff as ever Ginger.

Even in my younger more serious betting days, which are too far gone for me to remember, I couldn’t have had the patience for such a method. It’s good to see how you approach it.

[Gingertipster]

How did you stake in your more serious betting days, buckers?

When first starting to bet I bet same stakes every bet but found overall profit depended on solely my outsiders. Winning a hell of a lot more when outsiders won compared to medium and shorter odds bets.

So I changed and instead bet to win the same amount, but then found overall profit depended on solely my shorter odds bets. Because of losing a hell of a lot more when shorter priced horses lost.

Then tried returning the same amount, which was better – staking more on the shorter priced horses yet winning more on outsiders. eg £100 on an Even money shot, £!8.18 on a 10/1 shot, etc. But the gap in profit still wasn’t what I wanted.

When Rooster Booster won the Greatwood Hurdle under top weight, with a performance I immediately thought should make him a 20% fair 4/1 chance of winning the Champion Hurdle in March. I was amazed when Stan James offered 12/1 Had my usual bet at 12’s, put the phone down and immediately thought this type of value doesn’t come along often. So picked up the phone and had another chunk on… Few days later my Timeform Perspective came which said Rooster Booster’s performance would’ve been good enough to win three of the last four Champion Hurdles. 12/1 had long gone but went in again at 8/1…

Whole point of Rooster Booster was what massive value I considered him to be, illustrating how stakes (and therefore profit) should be higher the greater value is thought to be in the bet… As well as still allowing for what is considered its chance of winning.

Value Is Everything

[buckers]

Ginger, my more serious betting days were a long time ago. I’m sad to say that a big win would be £50 for me, that tells you how long ago it was + this would always be on course. More often than not York, Doncaster or Newcastle. Great days they were for me. I didn’t ever have an exact staking plan, I just wanted as we used to say the 50 notes.

The only sign of a staking plan that mirrors you would be that as often as I could I would bet each horse to win the same amount of money.

[tbracing]

Is your betting plan similar to Kelly, Ginge? Logically your staking plan makes sense, stronger the chance and more value the stronger the bet.

While I apply probabilities to all my bets, I have never been quite so comfortable with kelly. While I trust my prices I have to remember that it is an approximation. Kelly falls down for me in that I can price something 2/1 for a standard handicap say, and I will consider it a good chance, but I will find 2/1 shots I prefer to bet more strongly.

My bet plan is broken down simply to 100 points, I bet 1-4 points depending on price and value, so I am applying similar to you in that regard, Ginge. Generally if I price something 5/2 under I consider it a 2pt start bet and consider more on top. For a few occasions of the year I am prepared to risk 10% of my bank, the last one for me that was this was Baaeed.

A good way of growing your bank mind, Mark, is a progressive staking plan. When I first started growing my bet funds going forward, every time I was above my highest point I would re-adjust the point, it allows the bank to grow that bit quicker. So say you have 10k, you lose 500 but keep at £100 a bet unless you lose so much, but as soon as you are over 10k you move your stake so if you jump to 10.5k you would adjust your stake to £105 allowing you to always bet a little bit bigger each time and make extra gains.

I suppose there are many ways to stake, but managing a bank and staking plan is crucial.

[Gingertipster]

Sorry haven’t replied sooner tb, haven’t looked in to this part of late.

Yes, although I came up with the first version of this stronger chance / more value staking plan before I’d heard of Kelly, it does have many of Kelly’s principles.

Have since looked at Kelly but to my way of thinking the chances of going broke and / or losing a lot in a short time seem too high. Although appreciate you can do 1/2 or 1/4 or even 1/10th Kelly.

Yes, there are many ways to stake, but imo any successful staking plan needs similar principles. Your way of progressive staking also seems along the same line of thinking. Personally I like to have more than 1 to 4 (may be I need more because of a wider range of prices) but can certainly see your logic.

Value Is Everything

[tbracing]

No problem, Ginge.

I remember loosely your staking plan from the past, it certainly seems to work for you.

I have staked using Kelly and like you say can do 1/2 kelly 1/4 kelly etc to mitigate a bit of risk.

Staking is certainly something to examine closely as it is crucial, so thank you for explaining your method.

I have a simple kelly criterion calculator in a spreadsheet if anyone wants it.

TB

[Gingertipster]

EDITed out because there’s something I’ve found that’s even better…

Value Is Everything

[Gingertipster]

EDIT

Yes, I know.
Difficult to believe isn’t it?

Value Is Everything

[Gingertipster]

EDIT.

I’ve found Racing’s holy grail.

It’s taken time, but have now adapted my old Staking Plan to find a percentage of the bank to stake on each selection…

Drum roll please…

Value Is Everything

[Gingertipster]

A staking plan to rival Kelly!?
A staking plan especially for punters who often back more than one selection in a race / market.

GINGER’S CRITERION:
G = ((P – A) x 10) + P / 10

Where:
G = The percentage of betting bank to stake.
P = The Punter’s opinion of the Percentage chance of winning.
A = Available odds in percentage terms.

Or to put it another way:
The Punter’s opinion of the horse’s chance of winning in percentage terms…
Minus the Available Odds in percentage terms…
Times by 10…
Plus the Punter’s opinion of the percentage chance of winning…
Divided by 10.
Equals the Percentage of the betting bank to stake on this selection.

Timesing and dividing by 10 can be done very quickly by moving the decimal point or adding / subtracting a zero anyway. For anyone with just a little knowledge of mathematics can often work prices out in their head; which in my experience seldom works using Kelly.

Something thought by the punter to have a 50% (2.00) Evens chance of winning where 6/4 (2.5 or 1.5/1) 40% is the best price available…
50 – 40 = 10
10 x 10 = 100
100 + 50 = 150
150 ‘/, 10 = 15.00
15% of the betting bank to be staked @ 6/4 for this selection.

15% compared to what would in the same circumstances be 16.67% in the Kelly Criterion.

Value Is Everything

[Gingertipster]

Here are some percentages and odds with a comparison to what it would be in the same circumstances when using the often talked about and well respected (amongst serious gamblers) Kelly Criterion.

Ginger’s Criterion (GC) has something thought by the punter as having a 68% (fair 40/85) chance of winning should have these percentages of the betting bank staked on them at the particular prices:
4/7 11.20%, 8/13 12.90%, 4/6 14.80, 8/11 16.90, 4/5 19.20
Compared to Kelly Criterion (KC):
4/7 12.00%, 8/13 16.00, 4/6 20.00, 8/11 24.00, 4/5 28.00

The percentage of the bank staked when something is massive value is a lot bigger than what it is with lesser value – particularly on the shorties. 19.20% @ 4/5 compared to 11.20% @ 4/7. However, it isn’t as much of a difference as Kelly’s. imo No bad thing. One of the criticisms I’ve heard about Kelly is one loser can set the punter back too much. Because although short prices have a big chance of winning and maybe excellent value, it still might lose and (with Kelly) 28% is a massive chunk out of the betting bank in one go! My 19.2% would be bad enough. At even shorter prices the amount risked is obviously even greater. I tend not to bet at shorter than 4/5 anyway, not that there’s no value to be had there, just a personal preference. Just that I myself am better at distinguishing between what is a fair 2/1 and what is a fair 6/4… than I am between a fair 1/2 and a 4/6.

At 60% (fair 4/6) GC has the percentage of betting bank stakes as:
5/6 11.50, 10/11 13.60, 20/21 14.80, Evens 16.00, 21/20 17.20
Compared to KC:
5/6 12.00, 10/11 16.00, 20/21 18.00, Evens 20.00, 21/20 21.90

At odds-on I’ve included a wider “margin for error” than odds-against. Best to be as sure as possible something is a value (good) bet when wagering big money. e.g. A horse thought to have a 60% (fair 4/6) will not be backed @ 8/11 and in the vast majority of cases not as a main bet @ 4/5 either; but is bet @ 5/6. 60% to 5/6’s 54.5% being a 5.5% margin for error.

At 50% (fair Evens) GC:
11/10 7.40, 6/5 9.50, 5/4 10.6, 11/8 12.90, 6/4 15.00, 13/8 16.90
Compared to KC:
11/10 4.55, 6/5 8.33. 5/4 10.00, 11/8 13.64, 6/4 16.67, 13/8 19.23

I do consider backing something I believe a 50% (fair Evens) chance as a main bet at just two prices better 11/10… However, I’d seldom make it a “main bet” because main bets are those with the most value in the race. So at that price / value it is More likely to be a saver, getting back what I’d lose if the other bet/s in the race did not win.

When the form is a little bit difficult to work out I may reduce stakes particularly on marginal value bets because it is less probable my assessments are correct. ie Up to 25% less than these stakes. So something believed to have a 50% chance that is available @ 11/10 the 7.40% could be reduced by 25% (1.65) to 5.55%. If it is very difficult to work out then the race is thrown out – no bet. Conversely, if I believe the form is particularly easy to work out I may add up to 20%. So something thought a fair 50% chance available @ 6/1 would if (and only if) I believed the race particularly easy to work out could have up to an extra 3,00% for a total of 18.00%. However, the vast majority of bets will not be increased or deceased from these figures.

At 38% (fair 13/8 near as damn it) GC:
15/8 7.00, 2/1 8.50, 85/40 9.80, 9/4 11.00, 5/2 13.20, 11/4 15.10
Compared to KC:
15/8 4.93, 2/1 7.00, 85/40 8.82, 9/4 10.44, 5/2 13.20, 11/4 15.45

Shorties mostly have less of the percentage bank staked than Kelly. Again, imo Kelly risks too much when there is a massive amount of value. Ginger’s Criterion generally has stakes increasing at a slower rate too. This gap between Ginger and Kelly is largest the shorter the odds are. The gap getting smaller until a 38% chance available @ 5/2 actually has exactly the same (13.2%) staked under both Ginger’s and Kelly’s formulas.

At 33% (almost a fair 2/1) GC:
9/4 5.50, 5/2 7.70, 11/4 9.60, 3/1 11.30, 100/30 13.20, 7/2 14.10
Compared to KC:
9/4 3.22, 5/2 6.20, 11/4 8.64, 3/1 10.67, 100/30 12.90, 7/2 13.86

33% is the first time the sixth price has a bigger stake with Ginger’s Criterion than Kelly. ie 14.10% Ginger’s as opposed to 13.86 for Kelly’s.

At 25% (fair 3/1) GC:
7/2 5.30, 4/1 7.50, 9/2 9.30, 5/1 10.80, 11/2 12.10, 6/1 13.20
Compared to KC:
7/2 3.57, 4/1 6.25, 9/2 8.33, 5/1 10.00, 11/2 11.36, 6/1 12.50

At 16.5% (almost a fair 5/1) GC:
6/1 3.85, 13/2 4.85, 7/1, 5.65 15/2 6.45, 8/1 7.05, 17/2 7.65
Compared to KC:
6/1 2.58, 13/2 3.65, 7/1 4.57, 15/2 5.37, 8/1 6.06, 17/2 6.68

At 10% (fair 9/1) GC:
11/1 2.70, 12/1 3.30, 13/1 3.90, 14/1 4.30, 15/1 4.80, 16/1 5.10, 18/1 5.70, 20/1 6.20
Compared to KC:
11/1 1.82, 12/1 2.50. 13/1 3.08, 14/1 3.57, 15/1 4.00, 16/1 4.38, 18/1 5.00, 20/1 5.50

At 10% all these stakes are within 1% of Kelly… even by including eight prices.

At 6% (near nigh a fair 16/1) GC:
20/1 1.80, 22/1 2.30, 25/1 2.80, 28/1 3.20, 33/1 3.70. 40/1 4.20, 50/1 4.60, 66/1 5.10
Compared to KC:
20/1 1.30, 22/1 1.73, 25/1 2.24, 28/1 2.64, 33/1 3.15, 40/1 3.65, 50/1 4.12, 66/1 4.58

When stakes are small Ginger’s Criterion percentages of the bank are all a touch higher than Kelly. At 6% all these stakes are within 0.75% of Kelly. One of the criticisms I’ve heard about Kelly is the punter doesn’t win enough on a bigger priced horse. This goes a little way to change that – even though it will still (rightly) be less than many level stakes punters are used to.

At 2% (fair 50/1) GC:
80/1 1.00, 100/1 1.20, 132/1 1.45, 150/1 1.53, 200/1 1.70, 250/1 1.80, 300/1 1.90, 500/1 2.00
Compared to KC:
80/1 0.77, 100/1 1.02, 132/1 1.26, 150/1 1.35, 200/1 1.51, 250/1 1.61, 300/1 1.67, 500/1 1.80

At 2% the difference here is less than 0.25%, but at this kind of price a small difference goes a long way.

At 0.5% (fair 200/1) GC:
300/1 0.25, 400/1 0.30 500/1 0.35, 999/1 0.45
Compared to KC:
300/1 0.17, 400/1 0.25, 500/1 0.30, 999/1 0.40

…And at the biggest prices there is barely any difference.

I am a naturally cautious person and if a punter is more comfortable with smaller bets then – as with Kelly – it may be wise to use a fraction of Ginger’s Criterion. Half, Quarter, Eighth or even “Tenth Ginger’s” for the more cautious.

Value Is Everything

[Gingertipster]

The trouble I have with the Kelly Criterion is that betting more than one selection in a race is my thing… And although Kelly is an excellent staking plan for punters who take their betting seriously… imo It falls down when backing more than one horse / selection in a race / market. Particularly if they’re not put on at the same time.

Kelly Criterion says that for something believed a 50% Evens (or 1/1 or “2.0”) chance of winning that is available @ 6/4 (or 1.5/1 or “2.5”) a fair 40%, then the percentage of betting bank to stake should be 16.67%.

50 x 1.5 = 75
75 – 50 = 25
25 ‘/, 1.5 = 16.67
16.67%

To make it simple let’s use a betting bank of £10,000 so 16.67% of £10,000 (0.1667 x £10,000) is a £1667 stake. So under Kelly, £1667 @ 6/4 for a £2,500.50 Profit and a £4,167.50 Return.

So something believed a 50% chance of winning that is priced up as a 40% chance has 10% value in as much as (50 – 40 = 10).
And something with half of those odds and chances… Something believed a 25% fair 3/1 chance available @ 4/1 (fair 20%) has 5% value (25 – 20 = 5) but with Kelly:

25 x 4 = 100
100 – 75 = 25
25 ‘/, 4 = 6.25
Something thought a 25% chance available @ 4/1 gets 6.25% of the betting bank staked.

So to the same betting bank of £10,000 under the Kelly staking plan it’s a stake of £625 @ 4/1 that gives a £2,500.00 profit. So a single bet gives us pretty much exactly the same profit as the 50% Vs 40%, 10% value bet… Which on the face of it is fair enough. However…

Kelly is excellent when there’s only one selection in a race, but what if in another race my workings out show two horses as value bets? If two horses I have assessed as 25% chances are both available @ 4/1 (5.00) 20%… 2 x 25 = a combined chance of 50%… and 40% (2 x 20) being the combined price taken. The two combined being the exact same bet as the single 50% chance @ 6/4. But instead of an overall stake of £1667, Kelly gives us two bets of £625 for an overall stake of just £1250 (2 x £625) and combined profit on the race of only £1875.

One bet of £1667 stake for Profit of £2500.50 and Return of £4,167.50
One bet of £625 stake for Profit of £2500.00 for Return of £3,125.00
Two bets of £625 (2 x 625) = £1250 race stake for Profit of £1875.00 and Return of £3,125.00

I personally want any combined bet that adds up to a 50% chance at 6/4 with an overall 10% value to have the same overall stake and profit… whether I have backed one, two, three or more horses in the race. If the two 4/1 shots were available at the same moment in time there’d be less of a problem. Could add the two together 2 x 25 and 2 x 20 for a 50% chance @ 40% 6/4… Taking the £1667 and halving it to put £833.50 on each of the two horses. But often the pair are not available at the same time. So have devised my own “Ginger’s Criterion” Staking Plan.

Value Is Everything

[ep1987]

A simpler version is to set a point amount; let’s say £10 for example. Then just bet to win between 1 and 5 points depending on your perceived value; with a max stake of 5 points to mitigate exposure. With the most ‘juice’ usually being on bigger priced horses it pretty much takes care of itself and no complex calculations are required.

The amount you set your point amount at obviously depends on bankroll, risk tolerance and whether or not you bet every perceived edge or account for process risk and build in a buffer.

Personally I would advise setting a point at 2% of bankroll and erring on the side of caution with expected value; such that I tend to want horses that I think are at least 30% overpriced; as in I’d want at least 12s about a horse I thought should be 9s and would consider that a thin edge. The stake in that example would probably just be £1 with a bankroll of £500 and a point set at £10.

With other sports where there are less ‘unknowns’ it’s likely a higher volume approach with a point set at 1% would be more optimal.

[Author]

Posts