Monday, 23 May 2011

Hourly and Daily Winrates

Today, I took a look at a nice little feature HEM has for compiling your winrates and stats based on the time of day and day of the week under the Reports tab. I put together a bit of a spreadsheet based on that to see what time of day is best or worst for me. This would basically tell me when game quality is good and/or when I play my best. I found the results over a 300k hand sample for this year fairly significant.


A few observations for hourly:
Results are fairly terrible between 5pm - 11pm with the odd exception of 6pm - 7pm.
Decent, mediocre results from 8am - 10am and 12pm - 2pm.
Good results from 10am - 12pm and 2pm - 5pm.
Excellent results from 11pm - 3pm.
* Very small sample from 3am - 7am.

What I'm going to take away from the hourly:
  • Terrible results are probably mostly due to playing non-A-game and getting tilty after 5pm. Be very careful about playing this time frame.
  • Regs go for a supper break from 6pm-7pm?
  • Tables are not bad during the morning and early afternoon but not nearly as many fish as at peak time. Probably not a lot of regs awake either so it evens out a bit.
  • I probably play my best during the early afternoon until supper time.  Euros also getting home from work.
  • Play fewer tables to focus more playing against more regs.
  • Mass table weekends and maybe one afternoon session a couple days during the week.
  • * After midnight is skewed as shown below.

Observations based on day of the week:
It is true that I hate weekends because it's a variance minefield but I also make the most money by far on those days. At least now I have the data to prove it and can disregard my selective memory.
I don't know what it is about Tuesdays but it's probably the day that absolutely every fish is back to work and out of money and the games are at their deadest.
The rest of the week is kind of marginal but good enough to FPP pro against a lower Fish:Reg ratio.

* I generally work only weekends which explains the high 11pm-3am winrates. It also possibly explains some of my less than stellar results over the past year when I think back to how much money I used to make and my ability to play higher stakes when I wasn't working the weekend shift all the time. On Saturdays and Sundays I generally either don't play at all when I work a night shift and I play for a few hours after 11pm when I get home from the day shift. This is most likely skewing the after midnight hourly winrates because I tend to go to bed before 11pm fairly religiously during the week.

Actions based on day or the week:
  • Play as much as possible on the weekend in the short amount of time I have.
  • Quit sessions early and quickly during the week if they start off bad.
  • Put in more sessions in the 8am-4pm time slot skewed towards the tail end of that during the week to make sure I have a better chance of weekday A-game.
  • Take Tuesdays off to study or just plain take a break for the next while.
  • Figure out why I've played 2x as many hands on Mondays as any other day this year.
I'm going to stick to this for the next few months and then re-evaluate. I assume some of this is going to shift with the American player base now gone and everything might move forward a few hours.

    3 comments:

    1. Very interesting, tks. Although, I too am curious how Black Friday has effected things. Because, since the U.S. players were forced out, things have seemed a bit softer I think. YMMV.

      ReplyDelete
    2. what is the site and timezone for these results?

      ReplyDelete
    3. This is Stars time (EST).

      ReplyDelete