Betting Experts Answer: How to determine whether betting results are based on luck or skill? | Part 3

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

Over a small number of bets, luck can play a huge role in your betting success. Which is why it’s important to not just base your success on whether a bet won or lost, but to look further at other metrics. So, for question three, we asked the experts:

Q3: How do you determine whether your betting results were caused by skill or luck?

Check out Part 1 of our answers here, where we get the opinions of former odds compiler - Matthew Trenhaile, professional sports bettor - Spanky, host of the Business of Betting Podcast, betting blog - Day 25 and sports trader - Alex Ong.

Check out Part 2 of our answers here, where we get the opinions of football betting analyst - Mark O’Haire, tipster verification site - Smart Betting Club, founder of Punter2Pro, author - True Poker Joe and professional sports bettor - Jonas Gjelstad.


There is no 'best way', just a collection of things you can do, which when combined together, will paint a picture from which you will have to form a subjective judgement. How much you rely on each method will depend on circumstances, like what is your betting market and how big your results sample is. You will never perfectly know how much of your performance is skill and how much is luck, just that there will always be some of the latter, good or bad, although the larger your results sample is, the smaller the latter becomes via the law of large numbers. Four things you can do are:

  1. t-test / z-test - This tests how likely it is that what results you have, could have happened by chance assuming that there was no skill influencing those results. The smaller the probability that what you see could have happened randomly, the more confident you can be that it didn't happen randomly. This leap from probability to judgement is subjective. The test DOES NOT tell you what the probability is that you are skilled, just the probability that you can see what you see assuming you have no skill. My view is you want this probability to be less than 0.1% before you even start to consider it worthy of making any subjective judgement about your skill at all.
  2. Model versus reality - If you know what your prediction model says your results should look like, you can compare it to what they actually look like. If they look similar, or are not statistically far apart enough to cause you to worry, you can make the subjective judgement that your prediction model is doing what it is supposed to be doing. It's essentially the same method as 1) but from the opposite perspective. Plot the distribution of possible betting histories based on your model and see where your actual results history sits within that distribution.
  3. Closing line - If your model contains information that the market does not already know, theory predicts that the odds will move to accommodate it, after you have revealed it to the market. This means that after you bet, the odds should become shorter, if you possess skill. Theory also says that the most accurate odds are the closing odds. If you can systematically beat the closing odds you possess skill. If you beat them by more than the size of the margin, then you possess skill capable of making you a long term profit. Not everyone believes in the closing line value hypothesis. It is likely more reliable in larger, more efficient betting markets with more players and more liquidity. Its benefit comes from offering a much faster answer to the luck-skill question than methods 1) and 2). Statistical significance can be found with far fewer bets because the randomness in price movement is far smaller than the noise in actual results.
  4. Persistence -Does your performance repeat or does it wander like a drunken man. The former is a sign of skill, the latter a sign of luck. You want to see more persistence and less of a drunken man's walk in your profit growth.

Finally, there is one thing you should NEVER, NEVER use to answer this question: the size of your bank balance. That correlates to the size of the ego only, which is not a good measure of betting skill. Humility and an acceptance that most of what you see in betting is noise and chance are much better predictors of success.

Follow @12Xpert on Twitter & check out the 12Xpert website here.


Sample size is key here - minimum 700 bets, probably 1000+ would be useful at a minimum. Also there are tools you can use such as ‘z score’ which assesses the % chance that the results are down to luck.

Dan Weston

Beating the closing line can be used as a tool as well. But this is quite nuanced with markets with bigger lead time to an event (e.g. football) has more time for the market to be efficient than say, tennis, which often has a 24 hour or less turnaround between matches (and bettors don’t know who will be playing who, and the effects of their previous match, until that time).

Follow @Tennisratings on Twitter & check out Dan’s Tennis Ratings website here.


There’s probably no better or easier way to track your punting performance than measuring your bets against the closing market lines or prices. Sometimes the exercise will seem futile when you stumble across isolated bets where, even after the event, you are still convinced you were on the right side of value, yet the selection went off at a bigger price than you backed it.

But the bigger the sample, the more reliable this method becomes in helping you to identify when your faith has been misplaced - and hopefully, over time, you will begin to see the patterns of your most common biases. Beware, though: it's not an exact science. There are plenty of Expected Goals models that consistently beat the market price only to break even in the long run, because nowadays xG models drive the market more than anything else.

However, it’s reasonable to assume you won’t make money if you continually back teams at shorter prices than they are when the game kicks off. Many punters know this but very few actually do the work to track their own bets, and you’ll be surprised how much more you learn about your process when you measure yourself against the market, rather than final scores and the profit and loss column.

Check out Mike’s Fox Punter website here.


Good question, probably one of the most difficult aspects of betting. For liquid football markets using Pinnacle's closing line is always going to be one of the best indicators. In terms of horse racing seeing if you can consistently beat the Betfair SP is going to be a good sign.

With in-play betting and betting into soft markets that aren't available with sharp bookmakers, tracking results is going to key. Variance can be pretty huge in betting so not getting carried away by short-term results is important. The larger sample size of bets you have the better really for analysing your results.

A good start is to run your stats through a p-value test. This can give you a good indication of whether your results could be purely based on luck or whether they have been achieved through having a definitive edge.

Follow @SmSportstrader on Twitter & check out the Smart Sports Trader blog here.


The strongest indicator of good betting results being caused by skill as opposed to luck is of course the CLV (closing line value). CLV is a very conservative measure of skill, meaning that if you have it, one can hardly argue that your results aren’t the result of skill.

Church of betting

Of course, the betting markets are not 100% efficient so it is possible that you make good (and “skillful”) even without having any CLV to show for it. In this case, there are statistical methods that check what the chances are that your positive ROI is not the result of luck alone – like a t-test of the observed results of a betting record. However, in order to be able to establish the presence of skill beyond a shadow of doubt, you would need a much longer track record as opposed to with CLV. Therefore, CLV is not the only way to test for skill in a betting record, but certainly the best and most reliable one.

Follow @ChurchOfBetting on Twitter & check out The Church of Betting website here.

Check out Part 1 of our answers here, where we get the opinions of former odds compiler - Matthew Trenhaile, professional sports bettor - Spanky, host of the Business of Betting Podcast, betting blog - Day 25 and sports trader - Alex Ong.

Check out Part 2 of our answers here, where we get the opinions of football betting analyst - Mark O’Haire, tipster verification site - Smart Betting Club, founder of Punter2Pro, author - True Poker Joe and professional sports bettor - Jonas Gjelstad.

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

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

  • 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.
  • Q4: Kelly criterion or flat staking: Which stake sizing strategy do you consider to be the best and why? 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.