This is what is wrong with SPM

#31MasterSplinterWPosted 7/24/2014 6:51:54 AM
using_printer posted...
Pancake, my calculations are good, and illustrate that SPM, in its present form, does not accurately reflect a player's value or liability. Toss out some numbers if you disagree, but keep your proofless one sentence responses to yourself, otherwise.

If a stockbroker makes 1,000,000 in a year while losing 999,999, he's made one quid. But his SPM would be one million, which is dumb.


I mostly agree with those that are saying that if you look at SPM, W/L, and K/D, you can have a pretty good idea, so they don't see a need to alter SPM, which is kind of the camp that I'm in.

I'd also like to mention that you're above comparison isn't really that good. I think a more accurate comparison would be some guy who makes a million dollars in a year and spends it all . Yeah maybe some of it is a waste, but a 1 million dollar earner is pumping so much into the economy that it totally worth it (plus if he bought a nice house and some nice cars then he still retains a lot of the value). Looking at it as 1 million - 1 million is wrong. CoD is not a zero sum game, unless it's just DM or FFA, of course. And even then I still think it is hard to accurately evaluate what kind of impact (negatively/positively) a reckless/ aggressive player (but still good) may have on the more basic game types.
---
Most people are other people. Their thoughts are someone else's opinions, their lives a mimicry, their passions a quotation. (Oscar Wilde)
#32using_printer(Topic Creator)Posted 7/28/2014 5:18:53 PM
Splinter,

What about Kill Confirmed? Many times the top scorer on a team will be a player with substantially more tags than kills, while also have substantially more deaths than kills.

Which of these players would you prefer to have on your team:

A. 25 kills, 30 deaths, 30 confirms

B. 30 kills, 15 deaths, 12 confirms

Just assuming neither player called in a streak, and each player's kills came by gun, player A would have (50x25) + (30x100) = 4250 points.

Player B would have (50x30) + (12x100) = 2700 points.

If the game was 10 minutes, A is at 425/m while B is at 270/m. Yet player A's 15 more deaths has potentially awarded the opponent 1500 more points.

Player A is dying 3 times per minute with 2.5 kills per minute. Player B is killing 3 per minute, while dying 1.5 times per minute.

In this game, staying alive and getting kills equates to controlling the map, and putting your teammates in the advantage spots, making it easier to win. Also, the players with the greater K/D will always have a better chance of reaching a streak. The other more important inference is that Player A is depriving his teammates of streaks by virtue of having more tags than kills (IE stealing tags).

In my opinion, player B is more valuable than player A. Since there is so much interpretation that can go into judging SPM, it is presumably a broken statistic.
#33WorstSniperEverPosted 7/28/2014 5:29:56 PM
using_printer posted...
Splinter,

What about Kill Confirmed? Many times the top scorer on a team will be a player with substantially more tags than kills, while also have substantially more deaths than kills.

Which of these players would you prefer to have on your team:

A. 25 kills, 30 deaths, 30 confirms

B. 30 kills, 15 deaths, 12 confirms

Just assuming neither player called in a streak, and each player's kills came by gun, player A would have (50x25) + (30x100) = 4250 points.

Player B would have (50x30) + (12x100) = 2700 points.

If the game was 10 minutes, A is at 425/m while B is at 270/m. Yet player A's 15 more deaths has potentially awarded the opponent 1500 more points.

Player A is dying 3 times per minute with 2.5 kills per minute. Player B is killing 3 per minute, while dying 1.5 times per minute.

In this game, staying alive and getting kills equates to controlling the map, and putting your teammates in the advantage spots, making it easier to win. Also, the players with the greater K/D will always have a better chance of reaching a streak. The other more important inference is that Player A is depriving his teammates of streaks by virtue of having more tags than kills (IE stealing tags).

In my opinion, player B is more valuable than player A. Since there is so much interpretation that can go into judging SPM, it is presumably a broken statistic.


Player B all the way

Player A is helping the team win and helping the team lose, someone needs to compensate for him
Player B is just helping the team win, nobody needs to compensate for him
---
http://www.completegaming.net/forums/forum.php --- Come hang out with us!
ITT: WorstSniper is mad that he's spastic - flame030191
#34sycamotreePosted 7/29/2014 1:43:03 AM
In Dom (now that I'm decent) my SPM is ~350.
I play for streaks.

I usually go about 1.5-3 KDR, depending on opponents and teammates. (Obviously if teammates suck my performance will suffer, I'll explain if someone disagrees)

In a game I play fully I will never finish with less that 6 caps.

My strategy is to bypass B, actually. I take my homeflag and skirt B. I flank around and kill enemies trying to take B so that my team can get it. Once we do I just stay in the enemies face so they can't get to B (but I don't hover around B so I have few defends). I guess you could what I do "spawn trapping", but I do it by myself and if you can't stop one person you're bad. I just call it controlling spawns.

If my team can't get B, then I personally try to flip spawns. That helps my teammates get a new position to try from and gets me closer to streaks, and if I don't capture I've made the other team retreat and lose some of their map control to kill me.

Any good player knows that the easiest way to get streaks is to actually play the objective. I don't have to hold B, I've been in lobbies where I have A and C and B is neutral. Of course I only get on when I have control of spawns but I'm decent enough to know where people spawn. I just want to hold two flags and win.

I'm the guy who three caps a lot. My philosophy is that if we're doing well enough to three cap then spawns will flip but we'll still have map control, which means B control.

Goal is to get a VSAT, which is free map control, and ultimately a Swarm. (If i can)

I'm not a camper or a person who doesn't play the objective. I both play the objective and slay well enough.

Anyway, I tend to have less gun kills (if I got streaks) but the same KDR as a good player and the same caps as a obj who feeds. A game for me might look like.

30 kills (all gun)
15 deaths
8 captures
3 defends
1 VSAT (which means uncertain amount of assists points)

By your calcs, I'd have
3000-1500 = 1500
1500 + 1300 from caps (2 caps would be home caps on each side, worth 50 instead of 200) = 2800
Then it'd be 2800 + (10x), x being the number of VSAT assists. Let's say 10 for calc purposes. So 2900/10 minute games.

So a person with a 2.0 K/D and 8 caps who almost never let his opponents cross the map and gave his team a VSAT has a SPM of 290 in your system, which isn't bad but isn't extraordinary.

Mine is only as low as it is because of when I was bad (overall KD is only ~1.3 for similar reasons). So idk about this system. Seems like it nullifies my domination to eh status.
#35using_printer(Topic Creator)Posted 7/29/2014 3:56:47 AM
You forgot to multiply by your K/D, of 2.0 in that game, so it would be 290 x 2 = 580 SPM.
#36prsspardaPosted 7/29/2014 4:35:44 AM
While I agree , Running in like headless chickens constantly will not be beneficial , I have to disagree with your Higher K/D Logic, I usually play KC or TDM , So less strategy is required , However when I do play Dom or CTF , I usually go for the objective and not give a damn about my K/D. However We played a game and our top player was like 38/26 or something like that , I believe only two players had positive K/d's , I myself was in the negative but was playing the objective so did not really care.

They enemy had a top lad of 65 kills and about 6 deaths, did he have any captures ? Nope , Most of there team went positive but what was the capture count ? It was much more in our advantage and our team won comfortably enough.

What it boils down to is no one plays the objective ( Well not no one but very few ) To me that just kills the game type because if there is only you and one more playing the objective and the other 4 are too busy sniping or building up K/d's .

I never play them anymore unless I'm in a 4 man group or more because it is nearly pointless.
---
"Praise Be To Zanmato"
#37using_printer(Topic Creator)Posted 7/29/2014 12:56:08 PM
prssparad,

A player who is 65-6, in a 10 minute game of domination, kills 13 players every two minutes and dies 1 every two minutes. Generally, that is so much suppression of the enemy that it is the fault of his teammates for not exploiting map control and grabbing flags.

Also, I've had dozens of games in that range where I'll have 0 captures as well. I'll advance, clear out several enemies, jump on the flag but get killed by a 4rd or even 5th enemy before the flag is capped because teammates aren't in the area. But that's enough points to typically drop in a hellstorm and lightning strike for an additional 5 to 10 kills for a net of 8 to 15 kills in that one life.

Or, that player you're describing could have completely bypassed A at the beginning, cleared out B and then been killed while capping with other teammates, who finish off the capture.

All these different little holes you can point out demonstrate how subjective SPM is from game to game. 65-6 with 0 caps may be a game changer, or 25-12 with 8 caps may be a game changer. Who knows what each of those net in SPM. That's why SPM is a broken statistic.
#38ShinRaKnightPosted 7/29/2014 2:13:27 PM
A player who is 65-6, in a 10 minute game of domination, kills 13 players every two minutes and dies 1 every two minutes. Generally, that is so much suppression of the enemy that it is the fault of his teammates for not exploiting map control and grabbing flags.

Doesn't this contradict your original statement. Minus the deaths this is a player that was focusing on kills and go no caps. So he was clearing the way for his team mates.

So in your OP, the 50-40 player with no caps was essentially doing the same thing. He was killing the enemy and drawing their attention to allow his team mates to do the objective.
---
http://i27.photobucket.com/albums/c196/ShinRaKnight/ShinRaKnightSig.jpg
http://www.xboxlc.com/cards/newblack/Abyssal%20Shift.jpg
#39HamJabroniPosted 7/29/2014 3:48:16 PM
Those deaths could lead to streaks and could cause his teammates to lose. 40 kills for the enemy is a lot to give up.
---
"People die if they are killed."
-Shirou Emiya (Fate/Stay Night)
#40Undy_BPosted 7/30/2014 6:39:15 PM
using_printer posted...
All these different little holes you can point out demonstrate how subjective SPM is from game to game. 65-6 with 0 caps may be a game changer, or 25-12 with 8 caps may be a game changer. Who knows what each of those net in SPM. That's why SPM is a broken statistic.


It's pretty easy to figure that out in SPM. Each gun kill is 100 and each flag capture is 200. Streak kills are ~10 or something? Either way, both of those players seemed beneficial towards the team.

I don't understand why you argue that it's a "hole" or "broken statistic." The amount of activity you average in a minute is calculated and you get your SPM. Higher SPM means you're a more active player and generally better if your k/d is consistent with it.

feel free to add me: BADBearGaming
I'll play with you some time
---
BADBearGaming