Jump to content

Nacho Simulation Football League (Season 26 - Taco Bowl XXVI Posted!)


Recommended Posts

Still in shock after watching the defense step up. Going to be another gauntlet vs Hamilton

That front 7 caused us problems week 1 with 3 sacks, 2 FF & 4 stuffs given up. Only weakness might be if the secondary isn’t on their game.
 

My team has been altered as Ezra was replaced who gave up 2 sacks. I’m excited to see what this game brings!! 
 

Best of luck to the final 4 

  • Like 3
Link to comment
Share on other sites

1 minute ago, TL-TwoWinsAway said:

For all you people filling out ballots: there's a WR on the Wattsville roster that led the league in YPR while having zero drops in all of S26. Vote Cedric Tillman for your WR3 needs.

When is the deadline. I still gotta do Salsa?

Link to comment
Share on other sites

Found two bugs in the AFS code today, one major and one minor. First the funny minor one:

  • Kneels were not set up to check for safety. Normally this would never matter since the sim's AI would never select a kneel if on the 1 yard line, but the sim also has a play selector "game mode", so it could be forced. This was an easy fix... it now checks to see if the kneel results in a safety, and awards one to the defense if so. As this would never happen in the NSFL as the games are completely simulated by an AI designed to never select a kneel in that situation, the bug would have had no impact.

Now, for the other...

  • One of the methods responsible for decaying player attributes when they should be tired (just one, not all - there are a few other conditions where attributes are decayed) for receptions had a major bug. The intended method is by far the simplest of the decaying algorithms: determine (using the player's Endurance attribute) how many receptions a player can have before decaying the attribute (in this case, Acceleration), then when the reception count hits a multiple of that number (checked by testing if there's no remainder when dividing receptions by that number), decay the attribute. The issue? It incorrectly checks for the player's number of rushes instead of their receptions (I very likely copied and pasted this code directly from the similar method used for rushes, but missed updating the if statement's variable)! What does this mean? For one thing, it means that, since the remainder of 0 divided by anything is by definition 0, the attribute was simply always decayed by 1 for very single reception for non-Runningbacks (while it was only intended to do so every 2-5 or so, depending on their Endurance). And for Runningbacks (or Fullbacks, I guess), they never saw their Acceleration decayed from a reception except if their number of rushes equals a multiple of the decay number (which could be triggered several receptions in a row if they don't have a rush between them).

This is a big of an "aha" moment for me, as a long while ago when I was updating the sim engine (probably shortly after adding player decay in the first place), I was getting frustrated by passing stats being slightly lower than what I was expecting (and more evenly distributed among elite and not-so-elite receivers) and rushing stats being slightly higher. Eventually, I simply compensated by buffing/nerfing other formulas to produce what I felt was the realistic passing and rushing stats I was seeking (in particular, at one point I know I made Acceleration 'matter' far less, reducing the difference between high and low values).

Now, fixing the actual bug would be rather easy - all I have to do is update the condition to check for receptions as intended. The problem is, of course, the simulation engine for years has been tweaked and modified to account for this. Simply fixing it would make passing far stronger than intended, and nerf rushing at the same time since RBs would begin to be fairly decayed for receptions. Everything I've done since adding player attribute decay has been (unbeknownst to me) accounting for the fact that receivers always lose one point of Acceleration every time they catch the ball; elite WRs would be putting up huge numbers (with QBs also benefiting by extension) while receiving RBs would fair significantly worse. I do intend to fix this, but not for this season - I'll have to carefully think and tweak a ton of formulas to do it. Hopefully the bug will be fixed for Season 27.

In the end, I'm unsure how much impact this would have had in the NSFL as all the sim's formulas were set up to account for this fact. I think fixing it would increase the difference you'll see from big-play and "possession" receivers, at any rate (and of course receivers with high Endurance vs low Endurance, which previously only mattered for other attribute's decays). It will just take some time, and I thought some may find this amusing.

  • Like 5
Link to comment
Share on other sites

34 minutes ago, TL-TwoWinsAway said:

I need to get these over to Nach by noon on Thursday, so I'd appreciate having them early Thursday. (And, yep, you have not submitted a Salsa ballot.)

I put the question mark on the wrong sentence lol.

Link to comment
Share on other sites

43 minutes ago, TheKillerNacho said:

Everything I've done since adding player attribute decay has been (unbeknownst to me) accounting for the fact that receivers always lose one point of Acceleration every time they catch the ball; elite WRs would be putting up huge numbers (with QBs also benefiting by extension) while receiving RBs would fair significantly worse.

Me, over here with Ekeler as my key offensive player:

Glancing Staring GIF - Glancing Staring Focuing - Discover & Share GIFs

Can we make sure those changes are incremental to avoid absolutely destroying key players during one offseason?

Link to comment
Share on other sites

6 minutes ago, TL-TwoWinsAway said:

Me, over here with Ekeler as my key offensive player:

Glancing Staring GIF - Glancing Staring Focuing - Discover & Share GIFs

Can we make sure those changes are incremental to avoid absolutely destroying key players during one offseason?

Formulas would be readjusted such that while receiving backs would correctly be affected by bulk reception decay, their overall numbers would be affected minimally. In particular, restoring the intended difference between high and low Acceleration values will benefit them, especially early on in the game before that value is decayed.

The overall goal will probably be to make how things are now be the standard after a single Acceleration decay, with players being a bit better than they are currently before the first decay, but worse after the second.

  • Like 1
Link to comment
Share on other sites

Hey Salsa people: who should our QB be? It seems like it's between Cousins, Herbert and Tua, and, frankly, I'm not in love with any of them. Each is high-turnover, and I don't like our chances if we're turning the ball over against Rodgers or Mahomes.

Discuss.

Edited by TL-TwoWinsAway
Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
×
×
  • Create New...