honey_badger82, on 19 April 2018 - 09:03 AM, said:
So after reading through this thread and hearing all the "carry harder" and "55% + wr players are rare" comments I decided to do a study myself and record data from matches. The original plan was to do 20 battles a day in one tank only for 2 weeks and record the data after each battle. Well even though most battles lasted under 4 minutes it took 15 to 20 min per battle to collect the data I needed for my study. I ended up just doing 20 battles because right now I don't have that much time but what I found just confirmed what I already knew, it was more to prove to those that require the proof to see reason. At some point I am sure I will pick back up with my study.
Now Mondays for me are typically not good in terms of WR but this one went exceptionally well. It's typically Friday thru Sunday that I do the most damage to my own WR. Here is s link to the raw data I collected per match and my hasty synopsis at the end.
http://imgur.com/a/st5CbNR well you will have to copy and paste it since I can't get it to hyperlink.
Ultimately 80% of battles went to the team with the higher average winrate with relatively easy reasons why they didn't the other 20% of the time (either Yolo, good player in a tank he sucks in or best players getting smoked off the bat, i.e. salty pickle who I was elated to see on my team until he went poof)
55%+ WR players are not so rare and MM tends to keep these guys together (had very few toons)
Most unicums play meds and lights, very fast maneuverable tanks so it's easy to say "carry harder" while they hang back and vamp off the team and either pull the win off while the rest are in smoking ruins or get 6 on 1ed then complain the team sucks. It seems even at their level of play they sometimes can't count and realise reds 4 flanker outnumber our 2 or that our 4 heavies will never get up the hill in time to save them.
1st kill doesn't really mean much (one post on the forums says it determines a win 90% of the time) if 2 or 3 of your tanks get tore up to get that kill it is relatively meaningless.
Generally in my findings over not just these 20 matches but in all my play I find if there is a basic understanding of tactics in blitz as real life tactics have little meaning here then matches would be much more balanced. I think WG should publish their guides more often and not just the roles of a type of tank. Do a sidscrape guide, a when to flank and when not to flank guide etc. Every new player should get these in game notification center.
Flanking, yes it has been done without knowing where the enemy is before and worked but that is not the norm. As a combat arms soldier of 18yrs we generally only flank on a known enemy position. You MUST establish a battle line first then maneuver. Becasue if you run off blindly in the beginning you run into what is known as an ambush. Take it from someone who has been in an ambush for real, it' not cool and usually does not work out well. In WoT blitz it almost never does.
I know many more battles are needed to get real usable data, those who wish can extrapolate what they want from what I collected and if anyone wants to carry on if they have the time I can send you the excell format I used. I am rather interested in seeing this experiment done at T4 and 5, see how many seal clubbers are out there
Interesting results. As you say, 20 battles is a small sample, but I am always interested to see the results when people collect data.
I tried collecting data a couple months ago. I would screenshot the end of game results and then fill in my spreadsheet the next day at work. But it ended up being a massive undertaking and I quickly gave up.