Crusader and lightsaber aren't showing up on four guys 1 ball playoff stats (we had a 1st week bye and they guest played for other teams).

I am not showing up on the Playoff Stats page either.

I don't know if this is in place or not but I would like to have an all-time stat chart per player going back as far as possible. Also there are season stats and playoffs stats but what about doing those combined for those people fortunate enough to play in the playoffs.

My teams have the wrong stats. I think that all of us have apparently played more games than we have by 2, some of our games may have been reported twice. I should be on 9 games but I'm on 11 played and everyone else has an extra 2 as well.

There should be a career stats system so you can view your collected plants, medals, kills, betrayals, etc. Starting with last season's stats that we still have.

The stats page http://www.grifballhub.com/ggl-playerstats-fl2011/ only shows regular season games, not playoffs, and it doesn't display anybody who played less than 8 games. So upgrade suggestion: include all games, for all players. Maybe do a pagination sort of thing if it's too much to fit at once.

WARNING: WALL OF TEXT AHEAD

Alright, this is going to get really complicated really fast, and does somewhat require explanation as to why I think this would be an important thing to add to the stats pages. This idea does also require some minor additions of Halo stats into the site's pages. Whoever wrote the code for the charts, or whoever generally maintains them, would be able to help me with that. I am going to mention a lot of radical additions which, if necessary, would be added to a new page designed specifically for these stats.

First of all, additions to the general stat system include:
Time each person has held the ball
Total amount of time each person held the ball
Total amount of time a team's opponents held the ball

All of this can be found on bungie.net right now simply by hovering over a player's name, then adding those numbers up to a total number.

Also, I would need per game stats added to the page. Saying one person has 57 kills and another has 29 kills is biased toward the 57, even if they had played 1 or 2 more games. 29 kills divided by 1 game = 29 kills/game. 57 kills divided by 2 games = 28.5 K/G. Not too big of a difference per game, but that was just an example.

In case you haven't noticed already, I keep making "I need"s. This is because all of these stats are just building blocks to my overall plan: A way where we can find an overall value of each player and each team. In case the idea of just a few numbers that can determine a person's strengths intimidates you ("How will we get these numbers?"), don't worry, I have been working on a lot of these stats already.

I am ditching conventional gaming statistics like "K/D" and "W/L" (for the most part). Instead, I propose we include K/K+D. In the first two games of Grifball I have played this week, I got 57 kills and 59 deaths, something a lot of people would say is either a -2 spread or a .96 K/D. But those number mean nothing to me. What makes more sense is to treat this like a batting average in Baseball. Every "plate appearance" in Baseball is what I call an "encounter" in Grifball. Every time you meet with someone, and one of those people dies, that is an encounter. Using my same numbers, 57 kills, 59 deaths, I can look at that and say: 57+59=116. 116 encounters. But what percentage of those encounters did I come out victorious? 57/116=.491 49.1% of times I met with my opponent, I won. That is less than 50%... I should fix that! See how that works, rather than saying you have a 1:1 K/D? It makes more sense that way!

After I figured this out, I started looking further into it. In Baseball, there is a stat known as slugging, which looks something like this: (1xSingles+2xDoubles+3xTriples+4xHomeruns)/AB. So they are awarded 1 point for each single they get, 2 for each double, etc. then that total is divided by at-bats. But Grifball is a totally different sport, so I adjusted that. I looked at it like this: (2xDouble Kills+Triple Kills+Overkills, etc.)/K That number I get will then tell me what percentage of my kills (not encounters) were multikills. Also, you can get 20 homeruns and 0 singles, but cannot get 0 double kills and 20 killtrocities. This is important because multikills are EXTREMELY important in Grifball, as I'm sure most of you know, because if you can get an overkill, you are more likely to help your team score than if you got 4 kills all spread out. In case you want to know, my MTI (What I'm calling this number) is .439.

I could continue telling you all about my statistics, but I would like to keep the formulas to myself unless an administrator wants to pick one of them up for the site. I have stats for a lot of things, including a stat that will measure offensive, defensive, and overall values for each player.

Now, what you are probably wondering now is what value these numbers could have to the game of Grifball. I see tons of potential for not necessarily my stats but this way of thinking on these forums. Obviously, we will still keep traditional stats, that is why I suggested a new page just for these (if this idea goes through). Right now, I try to look at the teams I am going to face and I see wins, losses, kills, goals, and medals. But how does that help me know what I am up against? I have 57 kills but he has 100, I feel threatened, but if I looked at the K/E and saw he had a .358 and I had a .491, I'd feel a little better about myself! These numbers can then be applied to a whole team as an average and can help match teams against other teams of similar skill rather than how long they've played the game.

Grifball is a sport, as well as all of gaming is a sport. Why give players in this game arbitrary numbers rather than hardcore averages and statistics? Could you imagine watching Baseball and seeing one player has 100 hits when another has 10? Well, what if the guy with 100 hits had 400 at bats, when the guy with 10 only had 20? That is a .250 BA vs. a .500 BA. One has played a lot longer and therefor accumulated more hits. Not necessarily that they are better, but they have more time in the game.

Now, don't get me wrong, I am not saying that these numbers are ABSOLUTE. If you have a .491 K/E you WILL kill someone every 49.1 encounters out of 100. Every player is, like the xkcd comic says, a weighted random number generator. One game we could get 0 kills, the next we could get 100. But overall, these numbers are what should be EXPECTED from each person.

I respect the staff's decision if they choose not to implement these stats, as they are a bit complex and may be a little too much effort for not a lot of reward in this game. This is just merely a suggestion, and I will try to have my stats for my team updated just to keep my own little record of who is doing well. So far, it is really looking smooth and is opening up my eyes to players on my team I hadn't thought would be that great until I saw these numbers and compared them to everyone else on my teams. My "Top 4" team switched around a bit, I can say that much, haha.

In conclusion, if any administrator actually made it through that giant wall of text and is considering using some of my findings, I would gladly send them to you in a private message and see what you think, except my formulas for overall defense and offense which I am keeping locked away for now. If anyone has any criticisms whatsoever, feel free to throw them at me. I appreciate any feedback this community has to offer me.

If there is anything I forgot, I will try to edit it in.

-Phoenix

That is awesome.

We used to have a stat very much like that (called Batting Average) back in 2008, on the old Grifball.com site. Incorporating something like that would be something we will consider.

As for bomb carry time, though, although it appears on Bungie, it's not a stat that we can automatically pull. I love that stat, too, but we're not going to be able to import it to our tables.

It took me a while and I was very confused about everything that included Baseball but I felt it made sense. However isn't there still the dreaded threat of the approaching decease of Bungie.net

Haha that is awesome, I never knew they had anything like that before. I just thought of it the other day after our two games and thought it would be a really nice addition to the stats page.

I figured bomb carry time would be difficult to get... damn. Oh well, it isn't that big of a deal. Now that I think of it, a Goals per Carry Time stat would be an interesting stat to look at.


Sorry, I made my explanation a little wordy.
Basically:
AVG: Kills / Kills+Deaths(Encounters) would be a sort of "batting average" or "kill average" rather than k/d.
MTI: (2*DoubleKills+TK+OK...)/Kills would be to determine what percentage of your kills are part of a multikill.
APM: AVG+MTI
KPG: Kills divided by goals, to determine how many kills a person gets per score

And that is all I really have without using carry time. Bungie transferring everything to Waypoint could hurt this, but I don't know how Waypoint runs their stats or how transfer will go between there and here. Hopefully everything comes out on the plus side for Grifball!

That is lovely. I like that. If you have any other formulas or better ways of documenting statistics, feel free to let me know. I would be very interested.
I'm pretty sure Kal was the one who programmed what you see in the stats page. He's cool.

Here is what my screen looks like on excel:


(In case you are wondering, I numbered everyone on my team to organize everything better, I am number 1)

All the formulas are plugged in to each box to automatically calculate each stat. Then I have a separate Word file that tells me the meaning of each stat and the formulas.

I have all the overall stats for my team, all the average stats (stat/games played), all the ball carry times and scores. I have a chart for individual possessions, team possession, opponent possession, total possession, team, opponent and total goals, etc. You will notice columns of OFF, DEF, and PAA. For OFF and DEF I have my super secret formulas that will give me a number (no limit, not necessarily out of 10 or anything). Then, I took the average of both of them(I'll call them... AVD and AVO), that is the numbers at the bottom of those charts. Then I did: (DEF-AVD)+(OFF-AVO) which gave me Points Above Average (PAA). That is my "Overall" score. If you believe all my statistic bullpucky, that should tell you how good a player is above the "average player." Obviously, I don't have enough statistics to clearly define an "average player," because I don't have the stats for EVERY player in the league, but that is the average for my team.

I don't know if this has been talked about (don't feel like browsing the previous 7 pages for it) but in my opinion the stats you gain as a fill in shouldn't count. Only the stats you gain while playing for the team you're registered on should be recorded. This way we would actually know who the league leaders are in stat categories, and not who's played the most fill in games.

example:
after week 1 NOPLEX is leading in kills, deaths, spread, sprees, and medals. He has played 5 games though, 3 more than his team has played. If you calculate his average in those categories against others averages, he would not be leading in some categories if it only displayed his 2 games from his team.

this is nothing against noplex, and nothing against letting people fill in, I just feel in order to keep track of true league leaders, the stats should only display your scheduled games.

i disagree. if its a ggl game it sould count. i would think different if there was somthing gained for having high stats but scince there isnt i dont see the point. obciously if you play more games than your stats will be higher.

It'd be difficult to keep track of "real" games vs. sub games, since the current system assigns a player to the first team they play on which might not be their actual team. But one way to take extra games into account would be to have an option to display the stats divided by games and/or minutes played, rather than just totals.