Premium Value Bets - Prospect theory and staking A decision based on the best possible information and evidence is a good decision no matter what the outcome – and often the result is a loss.


Prospect theory, developed by Daniel Kahnemann and Amos Tvervsky, points out that the negative effect of a loss is greater in one’s mental ”accounting” than the positive effect of a win of an equal amount. Even though a good decision is not dependent on the result, in practice wins and losses have an effect on every bettor’s mood. Is it possible to improve decision making in the future and minimize the effects of hindsight in the decision process through staking strategy?


”It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.”


Mark Twain has said:”It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.” The remark by Twain, contradictory in purpose, illustrates the mental requirements demanded of a professional sports bettor. A sports bettor has to be able to trust what he knows, but at the same time he has to tolerate uncertainty in the quality of his work. There is no meter to measure the quality of a sports bettor’s work in real time. The asymmetry of wins and losses assumed by prospect theory is an easy condition for every professional sports bettor to recognize.

Correct staking, meaning adequately moderate, is the single most important element for anyone aiming for profitable sports betting. Correct staking is even more important than first-rate probability estimations. Staking too high will eventually destroy anyone’s bankroll due to variance even if one’s probability estimations are completely accurate. Most sports bettors use Kelly’s formulasee our Betting Guide – or a derivative of the formula and they minimize their risk according to their risk profile, by using a divider for percentage shares calculated by the formula.

A very common topic for discussion among sports bettors is why the higher stakes and expected value recommended by Kelly’s formula are nowhere near as profitable as you would expect based on plain figures. There can be various explanations for every single event, such as biased analysis, match fixing or new information, which is not commonly known. One example of biased analysis is the difficulty of condensing a given team’s negative information into a mathematical model, because the accumulating difficulties for a team do not linearly raise the opponent’s winning probability.


“Frequent over analysis of betting events, due to significant losses, can lead to the weakening of one’s risk taking ability.”


High stakes and significant irregularities in the betting market are suspicious conditions for a sports bettor who has lost a bet. Deep investigation is absolutely rational and even necessary, but in the worst cases it may lead to conclusions and rectification efforts too fast. You can’t always find a rational explanation for market moves. According to prospect theory a high stake and a resulting loss will cause a greater mental wound. Frequent over analysis of betting events, due to significant losses, can lead to the weakening of one’s risk taking ability. Decision-making consequently becomes a fearful process due to poor results. Very low limits on maximum stakes, prevents this kind of bias. In this case staking can approach an ”even stake strategy”, which is not mathematically optimal, but as demonstrated in practice, is a reasonable option.

Another option is to take into account the mean probability estimation of the betting market into one’s own probability estimation on which the betting decisions are based. This method in practice improves the probability estimations of every analyst, as proved by research. Mixing one’s own figures with the figures of the rapidly moving betting market however, disturbs the clarity of estimate traceability and it also weakens the development of intuition. When it comes to learning it’s sometimes good to take a stand, to be wrong and to receive some blows to one’s ego.

Moderate staking prevents hindsight biases, where one’s own argument or theory is adjusted to events or new information after the fact. ”Even stake strategy” might feel too simple per se, but if you take into consideration the unrealized profit from the best betting events and combine this with potentially improved decision-making, based on even staking, this model can be quite useful, at least on losing streaks.