Create Account

Exploring Bonus TPE Based on Ice Time
#1

EXPLORING BONUS TPE BASED ON ICE TIME
-Toirekari

1.0 BACKGROUND
With the current system, top TPE earning rookies grind through the season and reach about 250 TPE by the end of the regular season, breaking the cap in their first offseason. Conventional wisdom holds that with so many capped players in the league, a sub-250 TPE player is simply not useful on the roster. With most rookies spending the bulk of their rookie season sub-250 TPE, it leaves SMJHL GMs in a tough spot. Do they give the rookies reduced ice time and concentrate on winning games and cups, or do they give the rookies lots of playing time in the hope that it will entice them to stay active?

The role of the SMJHL has been a long-running debate on the forums, for good reason. It is by definition a development league, to soften the blow of getting to the SHL and provide a platform upon which to learn the ropes. But of course, it’s no fun to track stats if they don’t mean anything and losing sucks no matter what.

My personal opinion is that from what I can tell, most people here have a pretty competitive spirit and forcing people to play ‘just for fun’ is a pointless task. With that in mind, the problem shifts from debating if the SMJHL should be a strictly development league to how to play competitively without sacrificing rookie ice time.

In real life, the only way to develop a prospect is to give him ice time. GMs balance the risk of worse short-term results with the reward of more skilled players, but in the SMJHL all player development is on an essentially linear scale tied to strictly off-ice activities.

My proposal is to award bonus TPE to rookies sub-300 TPE based on minutes played each game. This would give SMJHL GMs another parameter to balance, because playing the rookies will directly make their team better instead of indirectly doing so from the hard-to-measure increased retention brought on by having stats.

Side note: the league would likely have to switch the post-season away from everyone making the playoffs in order to provide a punishment for over-playing rookies and subsequently losing lots of games.

2.0 FORMULA
My proposed formula is this:

Bonus TPE = 0.01*[300-CURRENT TPE]*[MINUTES PLAYED]*[BASE BONUS]

Explanation of terms: 
• 0.01*[300-CURRENT TPE] is a term expressing how far from 300 TPE the player is
• MINUTES PLAYED is pretty self-explanatory. The amount of minutes the player played in the game
• BASE BONUS is a factor used to tune the possible values within desired parameters. I’ve selected 0.0334 as the BASE BONUS, more on that below

The Bonus TPE would only be able to be claimed for games played in that update cycle, so a player wouldn’t be able to go inactive and claim hundreds of TPE at the end of the season. This would also prevent tampering, because as Bonus TPE is collected and claimed, the subsequent amounts of Bonus TPE that the player will receive will go down.

[Image: Plot-1.png]
Plot 1: Bonus TPE per Game

A typical update cycle contains 14 games. I selected the BASE BONUS so that a player who plays 30 minutes a game at base TPE for a typical update cycle will receive 20 TPE.
Using the formula for Bonus TPE, I calculated how much TPE would be gained per game for players at different TPE levels and different ice times. The results can be seen in Plot 1. Plot 2 shows the corresponding values for a whole typical update cycle.

[Image: Plot-2.png]
Plot 2: Bonus TPE per Update Cycle

3.0 DISCUSSION
To see how it would affect the rookie season of a top-earner, I examined the updates of Elijah Jones of the Anchorage Armada and Ricky Spanish of the Halifax Raiders.

Plot 3 shows the real TPE production of Elijah Jones, as well as the adjusted TPE using the Bonus TPE formula. Jones averaged 15.62 minutes per game in his rookie season, but his GM may have played him more if ice time had a direct impact on his development. I have included both plots for comparison.

[Image: Plot-3.png]
Plot 3: Elijah Jones Theoretical TPE Gains

As we can see, the Bonus TPE doesn’t make a drastic impact on Jones’ TPE throughout the season. By the time the season started, he already had 225 TPE because of his creation date and therefore only would have received 13 TPE with his real ice time and 16 TPE with an increase to 20 minutes per game. For this reason, I selected Ricky Spanish for examination since he created much nearer to the draft.

Plot 4 shows the TPE gains of Ricky Spanish under the same conditions. Spanish had very similar ice time to Jones, averaging 15.7 minutes per game. I also included a theoretical value with an increase to 20 minutes per game for comparison.

As we can see, the Bonus TPE has a much more drastic effect on the earnings of Spanish. He crosses 250 TPE almost 2 update cycles earlier, and goes from 275 TPE to almost 300 TPE for playoffs. Over the course of the season, he would have earned 18 TPE with his real ice time and 22 TPE with an increase to 20 minutes per game.

It should be noted that neither player reached the cap earlier. Both still get to 350 TPE in their first offseason, when they purchase equipment and participate in Team Training Camp.

[Image: Plot-4.png]
Plot 4: Ricky Spanish Theoretical TPE Gains

Another thing that should be noted is that this system would close the gap between top-earners and more casual players. Based on the calculated values in Plot 2, someone who plays 30 minutes per game and claims no other TPE sources (unlikely) could still reach 221 TPE from Bonus TPE alone (66 Bonus TPE!).

4.0 CONCLUSIONS
In conclusion, awarding Bonus TPE based on ice time for SMJHL players with less than 300 TPE would be a great way to incentivize rookie ice-time and allow rookies to quickly rise to TPE levels at which they can be an asset for their team. Top earners might see as little as 10-20 total TPE from ice time, which makes little difference in the long run. However, it comes at a time when the difference between 240 and 250 TPE might be immense.

Thanks to @Toivo and @Whikadoodle for having players, I hope neither of you mind that I pillaged your update threads.

I’d love to have a discussion about this in the comments, I’m sure there are things I’ve overlooked. If anyone wants to check my work I can post my spreadsheet I guess.

Code:
1107 words

[Image: EsVSFns.png][Image: uzmDXiJ.png]
Sigs by @Carpy48 and @tweedledunn
Reply
#2

This looks like nerd shit.

[Image: Screen_Shot_2020-07-07_at_12.29.58_PM.png]
Reply
#3

08-13-2019, 08:29 PMfrithjofr Wrote: This looks like nerd shit.

You right, you right

[Image: EsVSFns.png][Image: uzmDXiJ.png]
Sigs by @Carpy48 and @tweedledunn
Reply
#4

Plots in Excel... Cmom man gotta use that top end graphing software.
Reply
#5

08-13-2019, 08:47 PMgolden_apricot Wrote: Plots in Excel... Cmom man gotta use that top end graphing software.

Normally I plot in MATLAB (still not top end, I know, but fuck you I love it) except that I was at work and there's no license

[Image: EsVSFns.png][Image: uzmDXiJ.png]
Sigs by @Carpy48 and @tweedledunn
Reply
#6

08-13-2019, 08:56 PMtoirekari Wrote:
08-13-2019, 08:47 PMgolden_apricot Wrote: Plots in Excel... Cmom man gotta use that top end graphing software.

Normally I plot in MATLAB (still not top end, I know, but fuck you I love it) except that I was at work and there's no license

Dude yikes, get off that MATLAB shit and get on that Python shit

[Image: doubtfulalpha.gif]
[Image: 0XJkcN5.png]
[Image: sN8N4xa.png][Image: 639861613880541184.png] Cal Juice [Image: 639861613880541184.png][Image: RyzkmSj.png]
[Image: Eo2nBCt.png] Tomas Zadina
[Image: snacnei.png] Brady McIntyre
[Image: ice-level.svg]
Reply
#7

I love unique articles like this. Good job dude!

Disagree though, the come up as a rookie is part of the experience. Rookies also do produce at a pretty good pace if given ice-time. For example, Montreal had Jazz last year produce very well and this year has Rykov at a near PPG pace so far.

[Image: O8HREwL.png]


Raptors Canada pride
Player Page - Update Page
Reply
#8

08-13-2019, 08:56 PMtoirekari Wrote:
08-13-2019, 08:47 PMgolden_apricot Wrote: Plots in Excel... Cmom man gotta use that top end graphing software.

Normally I plot in MATLAB (still not top end, I know, but fuck you I love it) except that I was at work and there's no license

:( nooooooooooo
Reply
#9
(This post was last modified: 08-14-2019, 08:01 AM by Tomasnz.)

Nice work dude.

Main issue I see is having any tpe earn tied to the actual SIM makes it feel a little more manipulatable.

I think what I'd like to see instead is bring in SHL training regimes (but not EQ) into the juniors. That would mean that high earning rookies (IE active rookies) really can progress faster during the season.

The flip side about that is that it maybe promotes a need to accelerate that discourages people who are going to be midtier earners. (Who are still super valuable).

@"luketd" did a great article on how slow and steady earning (123/season iirc) can produce over 5 seasons about 1000 tpe. Which is more than enough for a good shl career.

http://simulationhockey.com/showthread.php?tid=94380

Definitely don't know how to solve for an issue where we have so many capped players playing 3 seasons in the juniors (hey I was guilty of this). But I think that is the issue. If roster balance was more like 1/3 capped players 1/3 250-350 and 1/3 rookies I feel like that would allow the rookies to contribute more.

[Image: tomasnz.gif]



Player Page
Reply
#10

08-14-2019, 07:57 AMTomasnz Wrote: Nice work dude.

Main issue I see is having any tpe earn tied to the actual SIM makes it feel a little more manipulatable.

I think what I'd like to see instead is bring in SHL training regimes (but not EQ) into the juniors. That would mean that high earning rookies (IE active rookies) really can progress faster during the season.

The flip side about that is that it maybe promotes a need to accelerate that discourages people who are going to be midtier earners. (Who are still super valuable).

@"luketd" did a great article on how slow and steady earning (123/season iirc) can produce over 5 seasons about 1000 tpe. Which is more than enough for a good shl career.

Definitely don't know how to solve for an issue where we have so many capped players playing 3 seasons in the juniors (hey I was guilty of this). But I think that is the issue. If roster balance was more like 1/3 capped players 1/3 250-350 and 1/3 rookies I feel like that would allow the rookies to contribute more.

Only need 78 tpe/season to hit 1k before you hit regression. If you earn over 100 a season it will be more. I’m actually working on a tool to show that stuff in depth

[Image: 0XJkcN5.png]
Czechoslovakia PROFILE || UPDATE || RAGE. Rage 
[Image: luketd.gif]




Reply
#11

08-14-2019, 08:01 AMluketd Wrote: Only need 78 tpe/season to hit 1k before you hit regression. If you earn over 100 a season it will be more. I’m actually working on a tool to show that stuff in depth

Huh, that's interesting. Is 1k considered a respectable level in the SHL for mainstay roster players?

----------

This is a great article, thanks for putting the work in.

[Image: 9ZNnX19.png]


Canada | Player Page | Grizzlies | Player Updates | Inferno
Reply
#12

08-16-2019, 06:51 PMcharlieconway Wrote:
08-14-2019, 08:01 AMluketd Wrote: Only need 78 tpe/season to hit 1k before you hit regression. If you earn over 100 a season it will be more. I’m actually working on a tool to show that stuff in depth

Huh, that's interesting. Is 1k considered a respectable level in the SHL for mainstay roster players?

----------

This is a great article, thanks for putting the work in.

Yeah I’d say 1k would be an average top 6 player. The thing is that if you hit 1k right before regression, then regression hits and you’ll be down to 900. So the general rule is to get all the tpe you can as well as fight regression(like earning tpe to mitigate regression). So like people who are active tend to be in the range of 1k -1.8k tpe. Then the super active hit more than that

[Image: 0XJkcN5.png]
Czechoslovakia PROFILE || UPDATE || RAGE. Rage 
[Image: luketd.gif]




Reply




Users browsing this thread:
1 Guest(s)




Navigation

 

Extra Menu

 

About us

The Simulation Hockey League is a free online forums based sim league where you create your own fantasy hockey player. Join today and create your player, become a GM, get drafted, sign contracts, make trades and compete against hundreds of players from around the world.