FanGM

We may or may not need to adjust how playoff averages are done to improve the efficiency of the league (see comments here for more: http://www.fangm.com/sportstalk/viewtop ... f=7&t=1185 )

Here's an example to consider...

Backup TE doesn't catch a lot of passes normally. Not catching passes means he doesn't get into most game box scores that we upload weekly. However for 2 games he came in due to injury and played a bigger role. He played 16 total games. Only 2 games did he show up in game stats with 4 catches each of those games (8 total).

Way we calculate averages now that's 8 catchers / 16 games = .5 catches/gm.

The only way to know he played 16 games, though, is to go back and look at everyone's total stats for the year. According to our already uploaded stats in the system each week, it would think he'd only played 2 games with stats uploaded. That's quicker for us to generate with data already in the system, but would view his average this way:

8 catches / 2 game updates in the system already = 4 catches/gm.

Another case: Star WR is suspended 8 games so only plays 8 total. He gets 800 yards on the season in those 8 games. Both the total game and number of updates in the system ways give the same results if he puts up stats each game. But one other option for us would be to use 16 games equally for everyone in that average calculation, no matter how many they play. So he goes from 100 yard average per game to a 50 yard average per game if we use 16 games for everyone instead of 8 games just for him. That hurts guys who get hurt a lot and players suspended, which arguably isn't the worst thing in the world to benefit the opposite kind of players who play every game.

We may not need to make changes to how we do things, but in case it's determined best for efficiency wanted to get some opinions.

I also hope we'll be able to keep the choice to use either real playoff games instead of average, but want to see how most feel on the importance of that. Since we continue to recognize injuries for players in the playoffs, it was in fairness that we also let teams use stellar playoff game updates. My aim would be to automate that choice if possible and have the systems keep track of managing real game or averaged and how many average playoff games used to determine the multi-game average for each round. If I can't get that all automated, though, may have to look into ways to simplify things on the manual labor front.

To clarify my votes, I like the option of having real playoff stats because it continues to draw in reality, but if it can't be automated by script, then I think we can drop the option.

I also like the way we do averages as long as it can be simplified or automated. If not, I would prefer to base it on the total number of updates in the system. I think the end result would be teams in the playoffs would be playing their best possible game, which may improve the quality of the games overall.

I would hate to get rid of the real game updates in the playoffs even if that means the season doesn't end until the middle of March. So I vote to keep everything the same.

Just a thought, but what about season averages divided by games played averaged with an average of games 13, 14, 15, and 16 divided by games played. That way there is somewhat of an advantage to a player being hot. Of course leaving out week 17 because clinching teams will be resting their starters and teams out of it will be playing rookies.


So......basically


season averages divided by games played

plus

averages of games 13, 14, 15, and 16 divided by games played during that stretch

divided by 2

= playoff averages

It's an interesting idea to factor in who's playing well at the end of the season and give them an edge.

May not help much with making that process more efficient or avoiding delays and would add stat preparation time to the mix possibly by adding in extra calculations and extra data to consider. But something to think about.

If we're going to offer the option of using real life playoff updates, then all the GMs should have to announce before the real life playoffs begin that they want them used for EVERY game reagardless of their next opponents in real life. The picking and choosing matchups isn't fair to me.

RyanM wrote:If we're going to offer the option of using real life playoff updates, then all the GMs should have to announce before the real life playoffs begin that they want them used for EVERY game reagardless of their next opponents in real life. The picking and choosing matchups isn't fair to me.

That's an interesting wrinkle to consider also. Originally the reason real playoff games began to be a choice was that sim teams with real NFL playoff performers were at a severe disadvantage to a sim team with guys at home on the couch (because guys still playing could get hurt and miss time where guys at home had their injury status frozen in time). And that didn't make sense to punish sim teams with lots of guys in the real playoffs. So we gave a benefit to teams with lots of real playoff performers to balance out the disadvantages to that.

With some of the manual labor difficulty in managing that, it's something I have to think more about and welcome input into the best ways ahead. Lots of options, just have to come up with the best for us.

to avoid 1 game wonders can it be season totals / game updates(minimum 8)

??

So game updates can not be less than 8. Instead of using 16 for everyone use between 8 - 16?

shouldn't be a big delay in the code adding a minimum 8 in there?

Royce R wrote:to avoid 1 game wonders can it be season totals / game updates(minimum 8)

??

So game updates can not be less than 8. Instead of using 16 for everyone use between 8 - 16?

shouldn't be a big delay in the code adding a minimum 8 in there?

That's an idea if things go that route to limit extreme cases.

Not voting yet, waiting for more discussion.