Does a rise in testosterone levels help us play better chess under pressure?

I've been reading "Performing Under Pressure" by Weisinger and Pawliw-Fry and I've found it extremely helpful for understanding what happens to my state of mind during a chess game and what I can do to improve my performance. However, one claim the authors seem to rely on in the chapter on confidence is the idea that an increase in testosterone is associated with better performance under pressure. They even cite this paper which studied sixteen male chess players at two OTB tournaments and found the following:

• those who performed better had higher testosterone levels the day before a regional tournament (but not before a weekly club tournament) as well as before the first game.
• Players who did better in each tournament had higher testosterone levels after each game.

Unfortunately, that study also has some clear flaws:

• There were only sixteen participants, and all of them were men
• The study did not control for rating of the players
• No causality was established
• As far as I could tell, this could also be caused by the 'winner effect', or the well-established idea that winning itself seems to increase testosterone levels.

I was curious to see how robust this finding was, and I came acrossthis review paper on the effects of testosterone in competition. They looked at over 50 previously published papers and drew the following conclusions:

• Competition causes a spike in testosterone levels for both men and women, although the sample size is smaller in women
• Winners have higher testosterone than losers in both men and women. For men, this was driven by an increase of testosterone levels during competition, whereas female winners tended to have higher pre-competition and post-competition levels
• Administering testosterone in women led to increased motivation for action and increased risk taking, but no studies established a causal effect between winning and losing.

Although most of these studies do not measure causality, the fact that other studies corroborate higher pre-competition testosterone levels hints at a possible causal link. However, other studies indicate that this could be caused by increased motivation, increased fitness level, social bonds between teammates, or even whether the competition is at home or on the road (see Casto & Edwards, 2016 for a review).

So, I conclude that our goal before an important chess game should not be to stimulate our own testosterone levels. That said, most chess players would agree that confidence is important for performing at our best, and several studies find that testosterone levels correlate with self-confidence, over-confidence, and 'competitive behavior' (again, see Casto & Edwards, 2016). So perhaps if we focus on strategies to increase our confidence, the neurological effects will follow.

References

Casto, K. V., & Edwards, D. A. (2016). Testosterone, cortisol, and human competition. Hormones and behavior, 82, 21-37.
Geniole, S. N., Bird, B. M., Ruddick, E. L., & Carré, J. M. (2017). Effects of competition outcome on testosterone concentrations in humans: An updated meta-analysis. Hormones and behavior, 92, 37-50.
Mazur, A., Booth, A., & Dabbs Jr, J. M. (1992). Testosterone and chess competition. Social Psychology Quarterly, 70-77.
Weisinger, H., & Pawliw-Fry, J. P. Performance Under Pressure.