|
|
Post by eric on Jul 28, 2018 18:23:39 GMT
100 150 points team 50 50 points
200 160 points team 200 55 points
500 170 points team 500 60 points
100 70 rebound team 50 20 rebounds
500 80 rebound team 500 25 rebounds
50 35 assists team 50 17 assists
200 40 assists team 200 20 assists
100 12 blocks team 500 22 assists
200 15 blocks team 200 7 blocks
500 17 blocks team 500 10 blocks
50 15 steals team 200 7 steals
500 20 steals team 500 10 steals
500 22 steals team 200 7 3
100 12 3 team 500 10 3
200 15 3 team 50 15 FTs
500 17 3 team 200 20 FTs
50 40 FTs team 500 22 FTs
200 45 FTs team 50 17 FGs
200 50 FTs team 100 20 FGs
200 22 FGs
100 20/20
500 5x5
50 Triple/Double
500 Quadruple/Double . . Okay, so here's what they used to be: 150 150 points team 75 45 points
250 160 points team 100 50 points
500 170 points team 150 60 points
75 70 rebound team 150 25 rebounds
250 80 rebound team 500 30 rebounds
75 35 assists team 75 17 assists
150 40 assists team 150 20 assists
100 20 blocks team 300 25 assists
200 25 blocks team 75 9 blocks
50 15 steals team 150 12 blocks
100 20 steals team 100 7 steals
300 25 steals team 250 10 steals
75 12 3 team 150 8 3
150 15 3 team 250 10 3
300 20 3 team 75 15 FTs
50 35 FTs team 150 20 FTs
100 40 FTs team 500 25 FTs
150 45 FTs team 25 15 FGs
250 50 FTs team 100 20 FGs
200 25 FGs
50 20/20
500 5x5
75 Triple/Double
500 Quadruple/Double And here's how often they happened in the past five years: n reward n reward
229 150 points team 703 45 points
46 160 points team 254 50 points
5 170 points team 18 60 points
213 70 rebound team 21 25 rebounds
5 80 rebound team 0 30 rebounds
462 35 assists team 272 17 assists
45 40 assists team 51 20 assists
1 20 blocks team 1 25 assists
0 25 blocks team 10 9 blocks
266 15 steals team 0 12 blocks
12 20 steals team 74 7 steals
1 25 steals team 2 10 steals
196 12 3 team 28 8 3
27 15 3 team 3 10 3
0 20 3 team 470 15 FTs
998 35 FTs team 48 20 FTs
331 40 FTs team 4 25 FTs
80 45 FTs team 2152 15 FGs
24 50 FTs team 175 20 FGs
4 25 FGs
230 20/20
1 5x5
305 Triple/Double
2 Quadruple/Double Rewards should be possible to get, so all those 0s have to go, and generate about the same bucks, so common ones should be worth less, rare ones worth more. With that in mind, I adjusted both what gave a reward and how much that reward was. Overall the average game in the old system generated 44.6 bucks, in the new system it's 44.3 bucks, so I think that's pretty solid. The new system (orange) is also a lot more coherent:  The dot all by itself waaaaay over on the right is 15 FG. Just too common, even at 25 dump bucks. I nudged the number requirements around a little, trying to stick with round or at least similar numbers where possible (hence all the 17s) then went with 50 bucks for any reward occurring as often as one in fifty games, 100 bucks for down to one in a hundred, 200 bucks for down to one in five hundred, and 500 bucks for everything else. Here's how it looks on a log scale:  You generally don't want your trend taking an abrupt right turn. Nice smooth lines, that's the ticket. . That's pretty much it, any thoughts let me know. If the league and Druce are down with it I can post it on the rewards thread and adjust the box score spreadsheet. |
|
|
|
Post by skrouse on Jul 28, 2018 21:14:16 GMT
I need to read through this and analyze it more. I think I mostly agree with the new one, but not entirely.
But I don't have time at the moment, FA won't be until next Tues or Wed right?
|
|
IanBoyd
New Member
Posts: 1,454
Likes: 315
Joined: February 2018
|
Post by IanBoyd on Jul 28, 2018 21:41:28 GMT
i'm fine with the new proposed ones
|
|
|
|
Post by eric on Jul 29, 2018 0:02:10 GMT
I need to read through this and analyze it more. I think I mostly agree with the new one, but not entirely. But I don't have time at the moment, FA won't be until next Tues or Wed right? day 1 is monday |
|
fason
New Member
Posts: 743
Likes: 291
Joined: February 2018
|
Post by fason on Jul 29, 2018 0:14:47 GMT
Didn’t read, but voted yes because I trust Eric.
|
|
Norman Dale
New Member
Posts: 449
Likes: 108
Joined: February 2018
|
Post by Norman Dale on Jul 29, 2018 6:53:36 GMT
Voted new
|
|
|
|
Post by TimPig on Jul 29, 2018 16:50:32 GMT
Voted yes for new.
Eric, it seemed like there were only a couple of suggestions for additional rewards. Would you be able to use the same data you pulled to see how often teams held other teams under certain scoring threshholds, and margins of victory?
|
|
|
|
Post by TimPig on Jul 30, 2018 18:27:27 GMT
eric assuming these wouldn't be implemented until next season, so can we run the box score macro today so people can start accepting/declining?
|
|
|
|
Post by eric on Jul 30, 2018 18:41:46 GMT
eric assuming these wouldn't be implemented until next season, so can we run the box score macro today so people can start accepting/declining? if it was up to me i'd implement them today. i also have your other requests on my to do list |
|
|
|
Post by eric on Jul 30, 2018 18:52:30 GMT
Voted yes for new. Eric, it seemed like there were only a couple of suggestions for additional rewards. Would you be able to use the same data you pulled to see how often teams held other teams under certain scoring threshholds, and margins of victory? margin of victory turns out to not work well with how my spreadsheet is set up, so i'm not going to do it. holding under a points threshold works well though, and here are the figures i've come up with: 100 bucks - 80 points alloed 200 bucks - 75 points allowed 500 bucks - 65 points allowed |
|
|
|
Post by skrouse on Jul 30, 2018 19:09:06 GMT
eric assuming these wouldn't be implemented until next season, so can we run the box score macro today so people can start accepting/declining? if it was up to me i'd implement them today. i also have your other requests on my to do list People already claimed rewards last season on the old scale, can't implement for 2004, would have to be 2005 forward.
Seems everyone is on board with the new changes but I feel like these changes will really hurt more balanced squads who lack a stat stuffing superstar.
For example, I had 4 20+ PPG scorers on my squad last season yet only hit the 45 point game mark 3 times last year, though I hit the 150+ as a team 7 times. Bumping the reward down from 75 to 50 seems reasonable, but moving the scale up doesn't IMO, that will reduce the spread of players that hits the mark.
I'd be interested to see how many different players hit some of these more common goals (FGs, FTs, Points). To see if there are a few outliers that are inflating the numbers. (I.e. Taurasi, Firsto, Bossert, Neon)
Eric can you run the macro based on both sets and see how it compares by team? I'm wondering if the teams with the top 5-7 players numbers would increase, while all others would decrease.
You're probably right and these changes will be good for the league, but I'd just like to take a deeper look before implementing it.
|
|
fason
New Member
Posts: 743
Likes: 291
Joined: February 2018
|
Post by fason on Jul 30, 2018 19:41:13 GMT
I took away my yes vote for now, I'll need to see Skrouse's thoughts on everything before revoting.
|
|
IanBoyd
New Member
Posts: 1,454
Likes: 315
Joined: February 2018
|
Post by IanBoyd on Jul 30, 2018 20:12:13 GMT
Voted yes for new. Eric, it seemed like there were only a couple of suggestions for additional rewards. Would you be able to use the same data you pulled to see how often teams held other teams under certain scoring threshholds, and margins of victory? margin of victory turns out to not work well with how my spreadsheet is set up, so i'm not going to do it. holding under a points threshold works well though, and here are the figures i've come up with: 100 bucks - 80 points alloed 200 bucks - 75 points allowed 500 bucks - 65 points allowed 75 and 65 points allowed for the most part isn't possible anymore. |
|
|
|
Post by TimPig on Jul 30, 2018 20:20:18 GMT
People already claimed rewards last season on the old scale, can't implement for 2004, would have to be 2005 forward. Good point that I didn't think about. |
|
|
|
Post by eric on Jul 30, 2018 20:35:28 GMT
margin of victory turns out to not work well with how my spreadsheet is set up, so i'm not going to do it. holding under a points threshold works well though, and here are the figures i've come up with: 100 bucks - 80 points alloed 200 bucks - 75 points allowed 500 bucks - 65 points allowed 75 and 65 points allowed for the most part isn't possible anymore. as a 500 buck reward 65 is supposed to be extremely rare, on par with a 5x5 or quadruple double six teams held the other team to 75 or fewer points in 2004 alone |
|
|
|
Post by eric on Jul 30, 2018 20:50:46 GMT
if it was up to me i'd implement them today. i also have your other requests on my to do list People already claimed rewards last season on the old scale, can't implement for 2004, would have to be 2005 forward. Seems everyone is on board with the new changes but I feel like these changes will really hurt more balanced squads who lack a stat stuffing superstar.
For example, I had 4 20+ PPG scorers on my squad last season yet only hit the 45 point game mark 3 times last year, though I hit the 150+ as a team 7 times. Bumping the reward down from 75 to 50 seems reasonable, but moving the scale up doesn't IMO, that will reduce the spread of players that hits the mark.
I'd be interested to see how many different players hit some of these more common goals (FGs, FTs, Points). To see if there are a few outliers that are inflating the numbers. (I.e. Taurasi, Firsto, Bossert, Neon) Eric can you run the macro based on both sets and see how it compares by team? I'm wondering if the teams with the top 5-7 players numbers would increase, while all others would decrease. You're probably right and these changes will be good for the league, but I'd just like to take a deeper look before implementing it.
i'll run it under the old and new systems for this year so we can see how teams change i won't include the points allowed rewards for this experiment new old change team
2550 2625 -75 Bucks
8550 5025 3525 Bullets
3750 4225 -475 Bulls
400 375 25 Cavaliers
650 1475 -825 Celtics
6550 5700 850 Clippers
4300 5900 -1600 Globetrotters
2000 3575 -1575 Grizzlies
4200 2800 1400 Hawks
4850 4075 775 Heat
1200 1675 -475 Hornets
4500 4075 425 Jazz
4250 5975 -1725 Kings
6950 8775 -1825 Knicks
1950 2750 -800 Lakers
5000 5675 -675 Magic
5200 7025 -1825 Mavericks
1950 2250 -300 Nuggets
9000 6450 2550 Pacers
2950 3275 -325 Pistons
1500 1425 75 Raptors
11150 10000 1150 Rockets
1400 1925 -525 Spurs
5250 5550 -300 Suns
5700 4550 1150 SuperSonics
0 0 0 team
6450 3600 2850 Timberwolves
5900 5150 750 Trailblazers
1400 1975 -575 Warriors
2850 3200 -350 Wizardsoverall the new system generated 1275 more dump bucks for the league, or about 44 per team |
|
|
|
Post by TimPig on Jul 30, 2018 20:52:33 GMT
Who on the Bullets is getting all those bucks
|
|
|
|
Post by Majic on Jul 30, 2018 20:59:38 GMT
Who on the Bullets is getting all those bucks this whats the big difference here? |
|
|
|
Post by skrouse on Jul 30, 2018 21:10:24 GMT
overall the new system generated 1275 more dump bucks for the league, or about 44 per team So overall more points were given out but to less teams.
12 went up 17 went down
Not quite as skewed as I thought it would be team wise but still, 57% of the increase went to 3 teams, which is along the lines of what I was worried about.
I think we need to revise what is being adjusted. Maybe add more tiers to points.
So instead of eliminating the 45 point tier, just add the 55 and adjust the rewards.
Instead of
75 - 45 Points
50 - 50 Points 200 - 55 Points 500 - 60 Points
Do
50 - 45 Points 75 - 50 Points 250 - 55 Points 500 - 60 Points
That's just 1 example, I don't have time to go through the rest, but we need to play with it a bit more IMO. Really appreciate all the analysis tho eric, you're one of the best when it comes to working toward improving the league.
|
|
|
|
Post by skrouse on Jul 30, 2018 21:11:14 GMT
(I will have time to go through this more tonight and the next 2 days though)
|
|
|
|
Post by eric on Jul 30, 2018 21:19:55 GMT
Who on the Bullets is getting all those bucks this whats the big difference here? the four biggest growers are Bullets Hawks Pacers Twolves these are also the teams that employed the top five shot blockers last year, and the twolves also had #7 |
|
|
|
Post by TimPig on Jul 30, 2018 21:20:44 GMT
this whats the big difference here? the four biggest growers are Bullets Hawks Pacers Twolves these are also the teams that employed the top five shot blockers last year, and the twolves also had #7 Yeah when I went and looked at the big jumpers, I noticed the Bullets and T-Wolves and thought of Simi and DJ Rony. |
|
|
|
Post by eric on Jul 30, 2018 21:28:51 GMT
overall the new system generated 1275 more dump bucks for the league, or about 44 per team So overall more points were given out but to less teams.
12 went up 17 went down Not quite as skewed as I thought it would be team wise but still, 57% of the increase went to 3 teams, which is along the lines of what I was worried about.
I think we need to revise what is being adjusted. Maybe add more tiers to points. So instead of eliminating the 45 point tier, just add the 55 and adjust the rewards.
Instead of
75 - 45 Points
50 - 50 Points 200 - 55 Points 500 - 60 Points Do
50 - 45 Points 75 - 50 Points 250 - 55 Points 500 - 60 Points That's just 1 example, I don't have time to go through the rest, but we need to play with it a bit more IMO. Really appreciate all the analysis tho eric, you're one of the best when it comes to working toward improving the league.
it's not a scoring issue though. the knicks and mavs had elite scorers in firsto picko and neon boudeaux and were tied for the biggest drop, meanwhile the pacers had taurasi and were big gainers, and the wiz had troy brown and just about broke even. there's no correlation. i also don't think a lot of teams taking -400 hits in a given year is a big issue. 17 to 12 sounds like a lot but another way of saying that is we're only 3 teams off from getting the split perfectly. |
|
|
|
Post by skrouse on Jul 30, 2018 22:36:04 GMT
So overall more points were given out but to less teams.
12 went up 17 went down Not quite as skewed as I thought it would be team wise but still, 57% of the increase went to 3 teams, which is along the lines of what I was worried about.
I think we need to revise what is being adjusted. Maybe add more tiers to points. So instead of eliminating the 45 point tier, just add the 55 and adjust the rewards.
Instead of
75 - 45 Points
50 - 50 Points 200 - 55 Points 500 - 60 Points Do
50 - 45 Points 75 - 50 Points 250 - 55 Points 500 - 60 Points That's just 1 example, I don't have time to go through the rest, but we need to play with it a bit more IMO. Really appreciate all the analysis tho eric, you're one of the best when it comes to working toward improving the league.
it's not a scoring issue though. the knicks and mavs had elite scorers in firsto picko and neon boudeaux and were tied for the biggest drop, meanwhile the pacers had taurasi and were big gainers, and the wiz had troy brown and just about broke even. there's no correlation. i also don't think a lot of teams taking -400 hits in a given year is a big issue. 17 to 12 sounds like a lot but another way of saying that is we're only 3 teams off from getting the split perfectly. Yeah points was just an example, may not be the best one. And 17-12 is good, but having 2/3s of the increase going to 4 teams isn't good imo. So we probably need to play with it more. |
|
|
|
Post by eric on Jul 31, 2018 17:43:26 GMT
it's not a scoring issue though. the knicks and mavs had elite scorers in firsto picko and neon boudeaux and were tied for the biggest drop, meanwhile the pacers had taurasi and were big gainers, and the wiz had troy brown and just about broke even. there's no correlation. i also don't think a lot of teams taking -400 hits in a given year is a big issue. 17 to 12 sounds like a lot but another way of saying that is we're only 3 teams off from getting the split perfectly. Yeah points was just an example, may not be the best one. And 17-12 is good, but having 2/3s of the increase going to 4 teams isn't good imo. So we probably need to play with it more. well, we really only have two choices: send most of the dump buck increase to shot blocking teams, or have dump buck inflation. while neither is ideal, i think the former is more palatable than the latter. the bulk of the bucks are always going to go to scoring overall since they get points for FG and FT and threes and PTS, but we're at least shifting it a little away from that. here are the %s of bucks awarded by category for the new and old systems: new old category
.519 .712 p
.119 .134 a
.076 .069 c
.085 .044 s
.095 .039 r
.106 .002 bwhere "c" is combo stats. i wasn't explicitly aiming for category parity but aiming for reward parity in general turns out to have worked pretty well for that. |
|
IanBoyd
New Member
Posts: 1,454
Likes: 315
Joined: February 2018
|
Post by IanBoyd on Aug 1, 2018 20:51:57 GMT
So are we good to move the rewards over to the new ones or nah?
|
|
|
|
Post by wee2dee on Aug 2, 2018 12:09:37 GMT
sounds like nah imo.
|
|
|
|
Post by Druce on Aug 2, 2018 12:37:33 GMT
well the new has 9 yes votes opposed to 0 on the old, and i don't particularly care either way, so if thats what the league wants then go for it
|
|
|
|
Post by wee2dee on Aug 2, 2018 12:43:45 GMT
just from reading the debate on the study in the thread, it seems it may need a bit more tweaking before being implemented.
|
|
|
|
Post by skrouse on Aug 2, 2018 12:46:13 GMT
Seems like most of the adjustments (in point spread*) are due to blocks, other categories are still distributed about the same rate, just up or down slightly.
I think we're good and can always reevaluate if we want to. Just going to be annoying to have to memorize the new minimums so I don't miss any
|
|
