Nice work on the spreadsheet and enjoyed your roundup even if i would disagree on a couple. However I have to point out that Ubers/min is a terrible metric. For a start your spreadsheet is including ubers farmed in set-up time (60% of season is stopwatch), which massively skews it against medics who don't consistently farm setup ubers (e.g. Buck). Even if you corrected your stats for this i don't think it would be a great metric. After all there are such large portions of the game where the optimal play is to avoid using uber. This cannot be said of the other stats: kills, damage, medpicks etc. So I don't think you can really use it equivalently. Personally I think Assists and KA/D are useful indicators for medic, although certainly less important than Deaths and probably heals too.

when u dodge half the season and still have the most soldier kills 8)) imo cykaz should get top spot for scout, but hey, this is all just a bit of fun anyway.

[...] The spreadsheet is nice, and I do agree with most classes. But especially for medic it doesn't make any sense. You go with Austin because his healing, but BC only got 28 hpm less, much better drop% stat (just because you use kritz you can't use that as an excuse to drop more: http://logs.tf/2165448), better assists per minute . And you said BC dies quite a bit, but he only got 0.04 D/M more than Austin. On top of that, it's much more difficult to get good stats on worse teams than on one of the best teams.

imagine caring this much about a couple of numbers on a google spreadsheet

[...] How dare you insult THE KING! This post/rating sucks quite a lot tbh but thanks for the spreadsheet. You can't rate players based on their stats alone when taking the whole season into account, especially when some teams didn't even have their mains on the classes on all match days. Mercs, subs, class changes mid season are going to affect the performance of the team, which can also affect how a single player performs. Plus maps matter too but the others already elaborated on that. You actually need to look at the single matches to see what actual plays happened and how they affected the outcome of the match, but that takes more effort than just looking at numbers. Rating players just on stats alone is a waste of time if it's so inconsistent in my opinion.

Just saying that you are treating as if they would obviously outperform everyone else even if they played the full season because coincidentally the games they did play they actually did well, therefore they will do the exact same performance on the other maps when this isn't always the case, certain maps are worse or better for certain classes and you also have to take inconsistency into the account aswell, you could be the best on koth but at the same time the worst on payload or just generally get bad luck and play a bad game out of the blue, this happens to anyone. The more games you play the more stuck your stats per minute or KD get, the more likely you are to make mistakes, say you get like a 2.0 KD on two officials and then hit a 0.5 on the other two, totaling an average 1.25 KD, now imagine another scenario, instead of playing the 4 officials and getting 1.25 KD, you literally only play the ones you get 2.0 KD, it paints a completely different story, you are just ignoring that and assuming "oh he played 2-3 games and did well, therefore he is better than someone who actually took the time to play this gamemode and had to keep consistency over a wide variety of maps. It's honestly hard to not find this insulting especially when it comes from the perspective of someone who doesn't even know what happened in any of the officials, it completely defeats the purpose of this thread since it spreads out false observations about players and discarting people who actually played.

[...] The spreadsheet is nice, and I do agree with most classes. But especially for medic it doesn't make any sense. You go with Austin because his healing, but BC only got 28 hpm less, much better drop% stat (just because you use kritz you can't use that as an excuse to drop more: http://logs.tf/2165448), better assists per minute . And you said BC dies quite a bit, but he only got 0.04 D/M more than Austin. On top of that, it's much more difficult to get good stats on worse teams than on one of the best teams.

Austin - In my premiership roundup, I forgot about a stat I care about somewhat - ubers per minute [includes non-medi-gun charges]. Spreijer got the most though, so it doesn't matter. However, it might have mattered here. No medic had stand out best stats across the board. Austin has the best healing, but drops over 8%. BC drops the least with good healing, but dies quite a bit. Buck dies the least, but has very few uber charges, and has poor healing. I decided to go with Austin, because the large healing amount, is very valuable, and his higher drop rate can be explained by nearly 40% kritz usage. However, in everyone's hearts, Buck has the best statistic of all. Buck is the only medic to confirm a frag on an enemy medic. Even better, he did it twice. Shoutout to Buck.