Hot or Not -- Week 2

[Sulla]

good write ups but big - u dun mention every1's stats

[milosh]

I like the split up of terr, spid, shark. How come team kills aren't a negative part of the shark rating? It's a lot easier to get kills when you also have 10-15 team kills. Does it mean you're better just because you bomb/mine a lot more?

[ogron]

I like the split up of terr, spid, shark. How come team kills aren't a negative part of the shark rating? It's a lot easier to get kills when you also have 10-15 team kills. Does it mean you're better just because you bomb/mine a lot more?

So, team kills are a part of the shark rating, it's just not calculated on the Main page of my sheet, it's in the secret sauce in the backroom. I'm still tinkering with the formula, but I currently have TKs divided by two as a negative. My reasoning is that shark mines at this level are generally on the enemy side, and the spider dying to it is partially responsible. Also your partner can intentionally die to your mine and it's a positive (and sometimes done on purpose), so I think 50% is fair.

Also hidden from view is that the formula checks your time won against your spiders performance. If your spiders were outkilled by 100, and you won the game or it was 15-13, it is making a qualifying judgement that you overperformed losing by only two minutes despite a lack of spider power, which is why you and Mark have done well in this metric, and Commodo and Mythril have not (they have few kills, no Teks, their spiders have slightly outkilled their opponents and yet they're -7 minutes total.)

You could make an argument against the formula, saying that spidering is a function of good sharking, and that conservative, responsible, low offensive output but quality reps is in fact improving your spiders performance and you're thus being punished for competent but un-flashy play. I wouldn't disagree with that statement. But they're stats, and I can't measure that. All the boxscores show are that theyre doing nothing offensively, their spiders are +, and they're negative on time.

Stats always need to be taken with a grain of salt.

It also heavily weighs kills per 30 minutes and Teks, and favors offensive sharks, simple because they're measurable. These stats are not a perfect encapsulation of sharking performance, but they're better than nothing, and at least giving sharks something to look at, since they always get the short end of the stick when it comes to this stuff.

The bottom line is if you're either winning time or getting Teks/kills, you're going to be putting up good stats. In the same way a spider may be putting up gaudy stats -- there is always the possibility it may be due to sub-optimal play (in the same way a spider camps for rec.)

I actually updated the latest shark stats to include your spiders kill differential per 30 minutes onto the main sheet before even reading this, and I think I'll incorporate it into terr as well going foward:

So, for instance, Markmrw's -65 means his teams spiders are getting outkilled by 65 every 30 minutes. A typical game is around 23-24 minutes, so around 50 kills a game. Yet he's only -2 minutes overall, has 5 Teks, and averages 15 kills a game and a 2.9 RPD -- from a statistical point of view that seems very good to me.

It doesn't matter what you do or what formula I use, though, if ph plays the majority of the season, he is the undisputed king of shark stats -- he can average 35 kills a game for an entire season easily

[Markmrw]

Shark STATS <3

[olde]

Imo being low on deaths when ur team is out, proves u just hide all game. It's like you are playing like Yojimbo then. U are not useful to your team and prob even missing out on easy vulchs. U dont even have to bomb/bullet all the time. It's all about pressure in jav.



@Ogron: I think u named Ro Roiwerk in the spid excelsheet.

it was a WB match u fkn goof!

[Server]

All the games are at the same time, so I have to mostly rely on the boxscores unless I catch the end of the match, hear about it on chat, or my team was part of it -- the NOTs are mostly there for entertainment -- not as an objective measure of who was truly the worst player in that game -- although they are often one and the same.

I wouldn't take it too seriously

Haha.. dont worry. I wont! Just posting for fun and messing around

[milosh]

Just getting the chance to read this, thanks for the well thought out reply