Actually, if you gathered a very large sample of statistics, you could create a baseline damage per heal value for each class, and then typify it by whether or not somebody performs above or below the baseline dph stat.

what about roamers vs pockets then

I have been noticing some patterns, where players from diverse leagues/divisions put up similar numbers to each other depending on their roles. It'd be sort of cool to gather a large sample of players and go from there.

If anyone can help me out with this, I'd greatly appreciate it. This may become something cool, and it may amount to nothing, but it's fun to try and find different ways to analyze the game.

I think it would only make sense if you could subtract total self damage from total heals received, that way the healing that has to occur to build uber/do rollouts would at least be accounted for.

EDIT: even this way is pretty weird since a roamer jumping around picking up packs could go into negative heals.

It also wouldn't take into account buffs which is pretty big. If a buffed scout kills an unbuffed scout you can't necessarily credit the buffed scout with being more efficient.

Yeah, we'd need to find a way to calculate that, or just only use the stat in relation to others in the same role, as Marxist mentioned. That way short leash pockets won't be adversely compared to health pack munching scouts.

I'm terrible at math, so any technical feedback is MORE than welcome!

As an example: http://sizzlingstats.com/stats/104347

Tonight as roamer I was a 1.06, my pocket was a 1.02, and our demo was 1.6 damage per heal.
We got beat, and comparing those stats to a winning effort like the Mix^ game seems to support the importance of the relative numbers.

I'd have to disagree with that, just because a buffed scout starting a fight is a good idea, and a non-buffed scout getting into said fight isn't a good idea, so the buffed scout deserves what he gets, and vice-versa.

I'm not sure exactly how beneficial this is; it's certainly something I consider (dpm vs. % of heals taken), but it's also very difficult to measure consistently. A team with a strong pocket soldier is going to make the pocket look very inefficient, even if he's making very effective use of those heals (drawing aggro for his team, making room for his flank, etc.) that won't be reflected purely by damage. Similarly, a team that pockets their star demoman will look like their demo is weaker than most, because he receives a lot more heals than other demomen.

You could make the argument that these are just results of these players being too heal-heavy and inefficient, but that's not how the game really works. Roamers and scouts rely on buffs to deal damage, which aren't reflected in healing statistics; this means that they'll look incredibly efficient, because all of their non-buff health comes from health packs, inflating their ratio.

It's definitely something you can compare between similar classes and roles; two aggressive scouts will have similar heal patterns, and can be comparable (if both scouts take 1000 heals but one does twice the damage, it's pretty clear they're being more efficient), but context is hugely important.

this argument doesn't make too much sense, as most stats we already use can't be used to accurately compare across different classes either

thought this was gonna be a thread about supplements

DDM is a pretty good measure of how well a combat class is doing. Sure, there are other factors involved, but the bottom line is how much damage people are doing. If you're doing less than 100, you're not hitting enough shots; if you're doing more than 300, you're probably playing really well - if you're a roamer or scout, it's really, really well.

If you then apply healing stats on top of those, you end up skewing the results. A pocket that does 300 DPM but takes up the majority of his team's heals will suddenly look like he's not doing well at all.

I could be wrong, but to me it just seems like there are too many factors to make this number mean very much, beyond which scout depends on heals more.

Not necessarily, if this was the case then I would see that the Pocket is seeing a lot of the fighting. And infer that this particular team is combo-centric in fighting style. Whereas if one of the scouts was doing above-average DDM and soaking up a majority of heals I would infer that the team has a stronger flank.

why does this not surprise me in the least
o that's right, just look at our team name
you dumb transylmaniac

OK SMAKA

heard wrong