That just means brav spends more time in games than actually in chat. So its not unusual to me that game time could be higher than chat time. For myself I bet that would be the case as well because I rarely spend more than 10 mins in chat waiting for a game. If nothing shows up, I just hang it up and leave. So if I spend 10 mins in chat but 20 in a game, then yea, my time logged in would be 10 mins. So the problem here isnt the stats, but rather the verbiage of what "Online time" actually is. I believe it's in chat time, and not a total time on the server.
Online time has only been recording since around Dec 4 2015, but min in game has been recording since about 19 July 2015
Min in game is highly accurate because it adds the minutes in the game report up, but min online is less accurate because it works by checking server status page every 3 minutes to see whos online, so its accurate plus or minus 6 minutes per session
Also min in game has 1 important flaw, if a game lasts over 1 hr it doesnt record the time in game because I havent programmed it to parse "1 day 2 hours 5 mins" it can only handle cases like "5mins" right now
So min in game and online time should always be compared separately. Also, online time includes min in game time. So if someone comes online for 1hr, and is in game for 30min, then 1hr gets added to the online clock and 30min gets added to in game clock
Min in game is a very accurate way to measure growth/decline of the server tho, it is a the green line on the server status charts, one of my favorite parts of the site is the server analytics charts
Btw, another big flaw in the system, like i said earlier, its not able to parse game reports for anyone not using English war2 client, because the game reports cant be understood