Betting Experts Answer: Betting staking strategy - Kelly Criterion or Flat Stake? | Part 3

Welcome to the fourth article in our series: “Betting Experts Answer The Industries Biggest Questions.”

This week we move onto one of the more contentious topics in the betting world; staking strategy and bankroll management, as we asked 11 experts the question:

Q4: Kelly criterion or flat staking: Which stake sizing strategy do you consider to be the best and why? And if Kelly bet, what percentage do you recommend?

Check out Part 1 of our answers here, where we get the opinions of football betting analyst - Mark O’Haire, betting analyst - Joseph Buchdahl, Smart Sports Trader - Ryan Bruno & author - True Poker Joe.

Check out Part 2 of our answers here, where we get the opinions of professional sports bettor - Spanky, host of the Business of Betting Podcast, betting blog - Day 25 and betting blog - The Church of Betting.


Inside betting

I have always advocated using a staking strategy that will not compound too aggressively any overestimation in edge and also feels comfortable to the bettor. The best staking strategy possible will only perform as well as the discipline of the bettor to follow it allows. If it makes you stake in a size you feel uneasy with then there is no shame in adapting it so that you can remain consistent and stay the course. Enforcing consistency is the most valuable part of a staking plan in my mind.

My personal preference for staking is a variable one with an edge estimation element to it. I like to first of all assess what I think a realistic margin would be for a given type of betting. A large historical sample is always nice but failing that I look at the margin the bookmaker bets the market to. So if the market is priced to 110%, I divide the bookmaker's additional margin by 4, so 10% / 4 = 2.5%. Similar in feel to using a quarter Kelly staking. The aim then is to place a bet that increases my bankroll by 2.5%.

So stake will vary with the size of odds but the EV remains constant for every bet. It is hard to achieve a level of accuracy in forecasting where you know the edge sufficiently to vary from one bet to another e.g. 1.7% for one then 2.6% for another etc. Making an estimation of superiority over a longer horizon without trying to over optimise individual bets is preferable to me.

The final layer is that in the previous example the 2.5% would be my max bet and I like to have three variations. I can have bets I calculate to be large EV but I have low confidence in the calculation and vice versa. The variations are my standard bet, my max bet (I can see clearly why the market differs from me and I know it is an area I am superior on forecasting on average than the market) and my FOMO (Fear Of Missing Out) bet for when I feel value is marginal but experience tells me that this given scenario has been a good one historically.

In the example these might manifest themselves as 0.5% FOMO, 1% standard and 2.5% max. It may look like the size relationship between the 3 is a little off to some but it has been adjusted to suit my temperament. Personal experience has shown me that my marginal bets over the years have actually not been much worse than my standard bets which I used to over estimate. While I always used to under stake my best bets due to caution so the increase from standard to max needs to be larger than from FOMO to standard.

I have frequently placed FOMO stakes on bets that I believe to be huge value from a model simply because those too good to be true value bets often are but I can't not have anything on it. If anyone were to adopt this method I would say feel free to use 4 or 5 sizes but I would suggest no more than 5. Any new betting venture for me historically has a simple spreadsheet where I enter the bankroll assigned to it and when I enter the margin of the market and the odds and my confidence level 1-3, it performs the simple calculation of what stake to use. Far from perfect but keeps me sane and staking less on things the less likely they are to happen a more when the EV is larger while not feeling like I am overestimating my own skills.

Bonus Knowledge: If for one reason or another you wanted your staking to appear outwardly more recreational then you need to try and invert the good behaviours of sensible staking. This means to be unsophisticated and inconsistent. When looking at very large data sets of bets the ones where the staking correlates with the probability of the event happening are more profitable than those which do not. Proportional bets are inherently smarter bets.

In the same way card counters should be wary of how large a spread they have between stakes when the count is positive and negative to avoid detection, so should the bettor looking to avoid detection. The stakes on 2.0 and 10.0 odds bets should be significantly different if EV is equal but by contrast the recreational bettor could quite easily have 1 unit on the 10.0 and 2 units on the 2.0. The recreational bettor's inconsistency manifests itself most obviously when chasing losses.

If you think you have a long term edge and can stomach the swings then increasing stake sizes erratically during a day of consistent losing will look like long term loser behaviour. Trying to do this without over-staking often means placing bets of a smaller than intended size initially so that you have room to look like you are increasing stakes similar to Martingale as you lose. Lost value at the beginning in order to reap greater rewards down the line is a risky business but in an era of big data analysis how you stake is as distinctive as a finger print for bookmakers looking to determine who has a clue and who does not. Very consistent 1 unit staking while suboptimal still exhibits a potential level of discipline higher than is desirable.

Check out Matthew’s podcast ‘Inside Betting’ here.



I don't recommend using the Kelly Criterion for a few reasons.

  1. The formula depends entirely on your ability to accurately work out the probabilities of future sports outcomes. Inaccurate estimates will result in incorrect stake sizes.
  2. The formula calculates extremely large stakes (in proportion to your bankroll). So you run a real risk of losing a huge % all at once.
  3. You're at the mercy of bookmakers and/or exchanges to accept those large stakes. It's unlikely that you'll be able to do so given account restrictions and limits, as well as the lack of liquidity for some selections.

In practice you're a lot better off using a level, or "flat", approach:

  • Level stake: You risk the same amount per bet, irrespective of the odds. So if the odds are 2.5 or 10.0, you’d still bet the same amount. For Lay bets you’d have to calculate the correct risk (the liability) per bet.
  • Level risk: You determine the risk from the odds. This method accounts for the implied chance of winning. Higher stakes are placed on low odds, and lower stakes at high odds. The result is an even profit from any win, no matter what the odds were.

Over time, the end result is more or less the same for both (1) and (2). I personally prefer equal stakes for its simplicity. I always reiterate that a betting strategy that fails to produce a profit on a level stake basis will not profit under any other staking plan - including Kelly. So firstly verify that your method is profitable (i.e. it identifies value bets) before you begin to optimise your approach to bankroll management. Staking plans will not 'save the day'.

Follow @Punter2Pro on Twitter & check out the Punter2Pro website here.

Check out Part 1 of our answers here, where we get the opinions of football betting analyst - Mark O’Haire, betting analyst - Joseph Buchdahl, Smart Sports Trader - Ryan Bruno & author - True Poker Joe.

Check out Part 2 of our answers here, where we get the opinions of professional sports bettor - Spanky, host of the Business of Betting Podcast, betting blog - Day 25 and betting blog - The Church of Betting.

Love getting the opinions from experts in the betting industry? Then subscribe to the Trademate Sports Podcast, where we interview the most important people in the sports betting industry.

Here are the other questions we have got our industry experts to answer so far:

  • Q1: Top 3 tips for betting beginners? Part 1 & Part 2.
  • Q2: How do you define “finding value” in betting markets? Part 1 & Part 2.
  • Q3: How do you determine whether your betting results are based on luck or skill? Part 1, Part 2 & Part 3.
  • Q5: What is the best method to use to make money from sports betting? Part 1 & Part 2.
  • Q6: How difficult is it to beat the sports betting markets? How efficient are the odds? Part 1, Part 2 & Part 3.
  • Q7: Best way to manage risk in sports betting? Part 1 & Part 2.
  • Q8: Is there a best sport to bet on? If so, what is it? Part 1, Part 2 & Part 3.
  • Q9: Assuming you have an edge, at what point can you start accurately evaluating your results and say that variance has played out? Part 1 & Part 2.
  • Q10: What is the one thing you would like to see change in the gambling industry? Part 1 & Part 2.
  • Q11: Do you think emotions play a part in people's sports betting results? If so, how should they overcome this? Part 1 & Part 2.
  • Q12: What are the top 3 mistakes people make when betting? Part 1, Part 2 & Part 3.

Get The Best Articles and News To Your Inbox

Subscribe to our newsletter and stay updated.