Confirmation Bias
One of the biggest traps that any punter can fall into is having and punting based on, mostly unconscious, Confirmation Bias and by Dog it’s difficult to shake. We can go to the interweb and find out the real definition it is,
“Confirmation bias is the tendency to search for, interpret, favor, and recall information in a way that confirms or supports one's prior beliefs or values. People display this bias when they select information that supports their views, ignoring contrary information, or when they interpret ambiguous evidence as supporting their existing attitudes”.
Remember that piece of toast that always fell on the floor buttered side down? It doesn’t.
And confirmation bias is one of the major reasons the punting 5S methodology tries to steer you in the direction of not listening to pundits or watching more panel shows on racing. I also say don’t read too many news papers especially during the Spring Carnival. You will be fed all types of tripe.
Simply,
Here’s how it tends to bite punters:
Remembering wins more than losses → If you backed a horse before and it won, your brain overweights that positive memory.
Looking for info that agrees with your pick → You’ll notice and value articles, tips, or stats that support your choice while ignoring contradictory ones.
Interpreting neutral info as positive → A horse’s “reasonable run” in its last race might be spun in your head as a “good sign” simply because you want it to be.
Overfitting past outcomes → You make a mental story — “Horse 7 runs well in the wet” — and keep betting it in the wet, even when recent data says otherwise.
Why it’s bad in racing:
Every race is a fresh set of conditions: distance, track, barrier draw, jockey, field strength, pace scenario, weather. Past wins under different conditions don’t necessarily translate to today.
Confirmation bias can make you double-down on losing bets, chasing the feeling of being “right” instead of following the value.
How to fight it:
Work from the data up, not the memory down — always assess the current race as if you’ve never seen the horse before.
Use numbers, not names — removes that emotional tag. That is the reason that in my race outcome suggestions that I use numbers. Something I have done for years now and it has improved my punting.
Write your reasoning before the result — keeps you honest about why you made the call. Insert comments on yout tissue calculator.
Track your bets & expected value (EV) — forces you to confront reality instead of selective memory. Keep records.
Actively seek disconfirming evidence — try to disprove your pick before you back it.
The roughie always win in small fields
So in the opinion of your humble author?
It is, absolute tosh about backing the roughie in a small field, absolute tosh
This is how confirmation bias works, when there is actually a roughie that wins in a small field, the radio race presenter, the television host, your mate having a punt with you and the Parish priest will all pop up with this pearl of wisdom, “the roughie always wins in a small field”.
Not only does it not always win, it rarely wins. In the almost 6 years I have been keeping records, not one outsider of the field has won in races consisting of 7 or less runners.
But in your (biased) mind it always happens because every Tom, Dick and Harriet tells you so and it convinces your brain it must be true. This is confirmation bias.
In days of yore, confirmation bias was called propaganda. In some countries it still is.
It’s (they’re) Due
What about “it’s due”. A jockey has had a run of outs or a trainer has had a run of outs and they are now riding or training a runner you like.
You place extra validity on the chances of your selection because your jockey or trainer “is due”. Rubbish!
The jockey may have been beaten in close finishes and/or ridden some great rides and come second many times so why is he due?
The trainers last 20 runners may have been contested by the 10 worst horses in his stable so why is he due?
Both are not due. They are no more likely to win this ride or train the horse in this race than even if they had both won their previous race.
The worst piece of confirmation bias about being due?
Don’t ever think that YOU are due. Don’t ever justify having a bet for this reason. The only thing that you can do when you think you “are due” is to be sure that you are due to have a spell from the punt.
Conversely, don’t ever place any credence on the fallacy “this jockey/trainer is flying” without doing some due diligence. A jockey or trainer in form can be a bonus to your selection but only if your selection has a chance anyway. An in form jockey cannot improve the performance of a horse that has no chance by enough to make it win.
The in form trainer cannot make his worse horse win just because his two or three best horses or well placed horses have been winning races.
This is “Recency Bias”.
One of the reasons that I have so many stables is that I do not want to rely on getting stuck with relying on one or two particular strategies to select winners. I want to back up any strategy with facts, facts that are contained in the data I collect and analyse. Even then, nothing is due. It is also the reason that you, like me, should record your bets and while analysing the results, give more time to the reasons some selections do not win rather than concentrating on the selections that win.
What distances do I have less success on?
What class of race do I have less success on?
What track do I have less success on?
Approximately 30% of favorites win and I say that it is better to look at it saying “70% of favorites lose”.
Critical thinking, self assessment and talking to other punters about your losses will help you more in your future punts than basking in the glory of previous victories.
These conversations are not conversations that will clutter your mind. They will help you continually apply red tags to information that you have come to accept in your previous punting phase. Here, we are continuing to apply Continuous Improvement.