An invitation for psychological study
11-12-2022
, 08:15 AM
I want to share with you some interesting information regarding the benefits of mental sessions for professional poker players. It is a big question whether mental training helps poker professionals. Since I’ve written all the dates when I conducted the private session with a player, I can easily track the actual results one month before and after mental training (MT). However, it is difficult to conclude that player results should depend on a mental coach's work. I am not aware of other specialists’ motivation in the field, but my method is based on providing the player with several useful tools which they can use to create a harmonious and fertile foundation around poker for themselves according to their needs and desires. Having said that, it is clear that if a poker professional managed to arrange a proper day-to-day routine, opportunity for self-learning, appropriate time to relax, physical activity and quality sleep, then we might expect to see their progress in the actual results.
I am relatively “new” in the field of mental support. Hence, only 66 samples were taken to collect players ROI one month before and after MT. In reality, I’ve conducted more sessions. But there were players whose statistics I was not able to track. Besides, I did this little research a month ago, and since then I’ve conducted a few more MT sessions. Also, it is important to notice that the number of 66 does not mean 66 players because some of the players came to more than one session.
The table with descriptive statistics is below.
ROI was computed manually through simple formula = PROFIT/(NTS * ABI + Rake) * 100
As we can see, the difference between before and after MT is not huge. First, however, let’s check the distribution of ROI scores which is essential for the statistical analysis.
PRE_ROI represent the ROI one month before MT, POST_ROI — one month after MT
Both graphs contain extreme values beyond normal distribution boundaries. For example, in before the MT condition, there was a case when one of the players won around 7000 in a one-month distance. Approximately 400 MTT tournaments are played during one month. Such a result would skew the distribution and spoil all the subsequent analyses. Thus, I’ve applied a filter to exclude the extreme ROI values from both before and after MT conditions. Below is the graph of the same parameters (PRE_ROI and POST_ROI) after the exclusion criteria was applied.
As you can see, now we have two perfectly normally distributed variables which we can analyse! I’ve applied the filter, which excludes all the results beyond the ROI range of -75 to 140. In total, 6 samples (10%) were excluded. The final sample of 60 results was used for statistical analysis.
To determine whether the difference between the two distributions is significant, I’ve conducted the t-test. When two normally distributed variables are compared, the assumption is that they come from the population distribution, meaning they are similar. P-value is the key to determining statistical differences in this test. P-value tells us the probability of obtaining the difference between two distributions, which we observe in the real world if assumed that these distributions come from the same population. As a rule of thumb, a p-value of 5% is needed to state that the difference is significant. In our analysis, we observe the p-values of 1% and 1.8% for ROI and Profit differences, respectively. Hence, I can conclude that in this sample of 60 results, the MT session significantly influenced the ROI and Profit.
But that’s not all! I can simply count the number of times when ROI was higher after than before MT without excluding the outliers. It gave 43 occasions (65%) when player ROI was higher after the MT session than before.
Obviously, the sample is too small to conclude that MT is necessary for a successful career of modern poker professionals. It would be interesting to combine the data with other mental coaches to investigate the benefits of MT on a larger scale. If somebody wants to share data, just let me know
Also, don’t forget to fill up the questionnaires for the experiment, which was mentioned in the first post!
The link to the experiment: https://essex.eu.qualtrics.com/jfe/f...hdXFYiNFTSyfjg
I am relatively “new” in the field of mental support. Hence, only 66 samples were taken to collect players ROI one month before and after MT. In reality, I’ve conducted more sessions. But there were players whose statistics I was not able to track. Besides, I did this little research a month ago, and since then I’ve conducted a few more MT sessions. Also, it is important to notice that the number of 66 does not mean 66 players because some of the players came to more than one session.
The table with descriptive statistics is below.
ROI was computed manually through simple formula = PROFIT/(NTS * ABI + Rake) * 100
As we can see, the difference between before and after MT is not huge. First, however, let’s check the distribution of ROI scores which is essential for the statistical analysis.
PRE_ROI represent the ROI one month before MT, POST_ROI — one month after MT
Both graphs contain extreme values beyond normal distribution boundaries. For example, in before the MT condition, there was a case when one of the players won around 7000 in a one-month distance. Approximately 400 MTT tournaments are played during one month. Such a result would skew the distribution and spoil all the subsequent analyses. Thus, I’ve applied a filter to exclude the extreme ROI values from both before and after MT conditions. Below is the graph of the same parameters (PRE_ROI and POST_ROI) after the exclusion criteria was applied.
As you can see, now we have two perfectly normally distributed variables which we can analyse! I’ve applied the filter, which excludes all the results beyond the ROI range of -75 to 140. In total, 6 samples (10%) were excluded. The final sample of 60 results was used for statistical analysis.
To determine whether the difference between the two distributions is significant, I’ve conducted the t-test. When two normally distributed variables are compared, the assumption is that they come from the population distribution, meaning they are similar. P-value is the key to determining statistical differences in this test. P-value tells us the probability of obtaining the difference between two distributions, which we observe in the real world if assumed that these distributions come from the same population. As a rule of thumb, a p-value of 5% is needed to state that the difference is significant. In our analysis, we observe the p-values of 1% and 1.8% for ROI and Profit differences, respectively. Hence, I can conclude that in this sample of 60 results, the MT session significantly influenced the ROI and Profit.
But that’s not all! I can simply count the number of times when ROI was higher after than before MT without excluding the outliers. It gave 43 occasions (65%) when player ROI was higher after the MT session than before.
Obviously, the sample is too small to conclude that MT is necessary for a successful career of modern poker professionals. It would be interesting to combine the data with other mental coaches to investigate the benefits of MT on a larger scale. If somebody wants to share data, just let me know
Also, don’t forget to fill up the questionnaires for the experiment, which was mentioned in the first post!
The link to the experiment: https://essex.eu.qualtrics.com/jfe/f...hdXFYiNFTSyfjg