[2024] Spring/Preseason Games NewSim Feedback

[alexfall862]

 

@Sarge including screenshot to help legibility of issue.

Ideas:
- Seems like maybe Khan caught it at the 1, fumbled on the 16, and it was a scoop and score for the Panthers but didn't get the defensive score captured correctly?
- Khan actually returned it 99, not 16 yards?
- IDK?!?!

@

[alexfall862]

I set Michael Carter's FB carries to 10, 

We ran 10 run plays from formations with a FB in them at he got 0 carries despite having a 2:1 advantage on distribution compared to the halfbacks, who split carries evenly across the game,

Not sure if this is an issue with Cater being listed as RB3 and getting a 0 on weights there (is it overwriting the FB position weighting for him?). Ideally I'd like to get him into a 20% RB carries and 100% FB carries situation, with around 15 carries a game, but the inputs seem completely borked atm. 

[tsweezy]

EDIT: upon further examination 9 of the 11 sacks happened after our starting LG was injured in the 3rd quarter. We had the backup set as the starting RG so I assume the sim just picked someone horrible like the punter to play instead (similar to how DTs are returning punts in blowouts). I guess I'm still curious who got selected but that would make a lot more sense than any of what I posited below (unless anyone else has an issue which I'll keep it up for)

-------------------------------------------------------------------------------------------

Speaking of sacks, can I get a refresher on how those are calculated OR a ballpark of how badly run blocking OL are hurt being in a pass blocking scheme?

These 3 defensive players combined for 9.5 sacks in our game

LE David Eck - 75 speed, 76 strength, 60 pass rush
RE Philip Wilcox - 83 speed, 84 strength, 88 pass rush
DT Lance Nguyen - 45 speed, 65 strength, 67 pass rush

I can see Wilcox having big numbers, but Eck and Nguyen should not be as good at getting to the quarterback as Wilcox, particularly when they're going against on OL of:

92 strength, 89 pass block
90 strength, 84 pass block
86 strength, 79 pass block
94 strength, 88 pass block
85 strength, 81 pass block.

Again, Nguyen and Eck having 6 sacks in 40 plays against that line seems crazy high to me. My best guess is that this was a hot fix to try and make long passes hurt more, but I'd like to know if that was the case or if there was just some crazy set up stacked maluses somehow, or if Morrison is just uniquely the worst player in the sim at not getting sacked.

Edited April 14 by tsweezy

[Bundy]

61	7	0	2	Minnesota Vikings	07:03		35 Minnesota Vikings	Kickoff	N/A	N/A	N/A	N/A	N/A	0	Man	Man	Man	85	K Mike Tsao kicks for 65 yards. Fielded at the goalline by RB Fish Liggins. RB Fish Liggins returns the ball 15 yards.
62	7	0	2	New England Patriots	06:55	1st and 10	20 New England Patriots	Pass	Quick	Double Tight	4-3 Heavy	Quick	Pass Defense	0	Man	Man	Man	0	QB Rashaan Lyons throws it... and it's incomplete. Pass intended for RB Jerry Lawson.

 

Not sure if the "fielded at the X" is just flavor but if its not, then wondering why a KR fielded at the goal line and returned 15 yards came out to the 20 (or is this an issue of the return yards being wildly off and it wasnt actually returned out of the end zone)

[Bundy]

85	7	0	3	Minnesota Vikings	12:22	2nd and 4	18 New England Patriots	Pass	Short	Singleback Gun	4-3 Base	Short	Run Defense	0	Man	Zone	Zone	3	QB Christian Sinnock throws to WR Adrian Daniels complete for 3 yards. Tackled by SS Johnny Nguyen. PENALTY: Offensive Holding. Offense. Player: OT Herman Knott. 0 yards.
86	7	0	3	Minnesota Vikings	12:05	2nd and 4	18 New England Patriots	Pass	Short	Spread	4-3 Base	Short	Run Defense	0	Man	Zone	Zone	5	PENALTY: Illegal Shift. Offense. Player: WR Alex Williams III. -5 yards.

 

Im sure this has come up before but logging it for reference, Holding didn't subtract yards, was just a replay of downs/ate some clock

 

EDIT:

but this other holding call did work

100	7	0	3	Minnesota Vikings	06:39	2nd and 5	20 New England Patriots	Run	Inside Run Left	Singleback Gun	4-3 Base	Inside Left	Pass Defense	1	Man	Zone	Zone	14	RB Yogi Oladipo carries for 14 yards. Tackled by ILB Ariel Calderon. PENALTY: Offensive Holding. Offense. Player: OG Mark Killough. -10 yards.
101	7	0	3	Minnesota Vikings	06:19	2nd and 15	30 New England Patriots	Pass	Play Action Short	Singleback	4-3 Base	Short	Pass Defense	1	Man	Zone	Zone	4	QB Christian Sinnock throws to WR Alex Williams III complete for 4 yards. Tackled by OLB John Damiano.

Edited April 14 by Bundy

[alexfall862]

Play 161 had both a kickoff and a 2pt conversion in the same play? Is this related to the issue with the Hue Khan 16 yd kickoff TD and the successful 2pt conversions counting at TDs?

This is play 161 of the Week 2 Rams/Bucs game.

[alexfall862]

Dunno if the other suggestion threads are supposed to be active - but I'd kill for a 4CB dime package in addition to my backup SS.

Base package has 4-3 and 4-3 over and it's a nice way to change things up, would be cool to have that for dime. If wrong channel, that's cool, just trying to drop it somewhere before I forget.

[Rocketcan]

1 hour ago, alexfall862 said:

Dunno if the other suggestion threads are supposed to be active - but I'd kill for a 4CB dime package in addition to my backup SS.

Base package has 4-3 and 4-3 over and it's a nice way to change things up, would be cool to have that for dime. If wrong channel, that's cool, just trying to drop it somewhere before I forget.

What scheme is this? Part of the reason some schemes have access to some formations over others is because users will need to make meaningful choices about what they actually want to do. No one scheme should give you everything you want as there is a give and take to everything. Otherwise one scheme would become the meta.

[alexfall862]

2 hours ago, Rocketcan said:

What scheme is this? Part of the reason some schemes have access to some formations over others is because users will need to make meaningful choices about what they actually want to do. No one scheme should give you everything you want as there is a give and take to everything. Otherwise one scheme would become the meta.

Speed scheme, though I don't think 4CBs instead of 3CBs and a backup safety is at all outside of the real of possibility for any scheme.

[Rocketcan]

36 minutes ago, alexfall862 said:

Speed scheme, though I don't think 4CBs instead of 3CBs and a backup safety is at all outside of the real of possibility for any scheme.

Again, the point is to force users to make decisions. Also, if you really want to run 4CBs, just put a CB at SS2.

[Rocketcan]

Responding to all of the feedback here since my last response.

1.

On 4/5/2024 at 8:46 AM, SageBow said:

It looks like the punt return fixes have been addressed but the math is not adding up. 

 

 

29 yard punt from the UTAH 34 -> BUCK 37

49 yard return from BUCK 37 -> UTAH 14

Actual -> BUCK 48

happens on at least all of my punt returns. Just noticed because the return yardage was massive.

Additional example

61 yard punt from UTAH 23 -> BUCK 16

81 yard return from BUCK 16 -> UTAH  3

Actual -> UTAH 31

Tuscan is working on an interface fix for this coming week.

2.

On 4/5/2024 at 9:06 AM, Piercewise1 said:

Purely cosmetic, but the sim is reporting field position as going all in the same direction. Here FAU starts a drive at their 19, but it's displayed as being on the Temple side of the field and counting up as they gain yards. So it's correct when Temple has the ball, but backwards for FAU. I confirmed this as happening in other games as well. Again, I don't think it affected anything in the actual game result (play call, 4th down decisions, etc) it's just displaying backward.

 

 

Also, while looking for a good screenshot to reflect the above, I found this phantom pass in the Temple/FAU game. Declared incomplete, but for -1 yards with two tacklers. Maybe the sim couldn't find the receiver correctly, so it recorded the result but called it an incomplete pass.

 

This should be another fix that is included this week on the interface. The second part of this piece of feedback will be addressed later in the post (point 21).

3.

On 4/5/2024 at 9:52 AM, dearden said:

Apologies if this one's already been addressed, though I couldn't find any thread about it and didn't notice this trend last week; it appears as though during Thursday night's Week 2 Spring Games that Quarterbacks were taking the field as two-way players for at least part of the game.

Within the "Defensive Stats" section of the Box Score, I'd noticed that my designated QB2, Geno Wimmers, made a solo tackle at some point during the game.  This was despite the fact that there was no indication that he had been on the field during an offensive turnover, as his "Passing Stats" indicated that he hadn't thrown an interception and our offense had not fumbled the ball.

Upon further investigation, I'd noticed within the "Play-by-Play" that he'd made the tackle in one of the last drives of the game on a 3rd and 10, indicating that he was probably just out there playing defense for whatever reason.  Being a relatively new coach, I'd figured that I'd messed up my Depth Chart somehow, but this pattern appears to have happened through all of Thursday night's matchups.  For example, within the other two games where a Quarterback did not throw an interception (Utah vs. Bucknell and Texas A&M vs. Northern Colorado), five Quarterbacks were listed under the "Defensive Stats" section of their respective Box Scores.  These were Utah QBs Jacob Dozier and Gerald Wright, Bucknell QB Jan Quinn, A&M QB Brian Bass, and Northern Colorado QB Orlando Dzubner.  While none of these Quarterbacks made a tackle, my immediate supposition is that, like my QB2, they were playing snaps on the defensive side of the ball for whatever reason at some point during the game.

Second round of apologies for how long-winded all-of-the-above was - not sure if I'm just misreading things or if there's a genuine goof somewhere, but I figured it was best to flag it ahead of the remaining games this weekend.

You did not goof. Backups were in, and sometimes that combined with injuries and stamina means the engine has to grab players from the bottom of lists to fill in spots. There will likely be another rework of this in the future, but this is not significantly detrimental since most of the time this is in blowout games and they tend to be players that aren't high on the depth chart on either side (read: guys you really care about).

4.

On 4/6/2024 at 2:59 PM, smackemz said:

Special Teams tackles are accounted for in the play by play, but aren't being tracked in the box score.  I think it would be cool to see those included in the box score. 

Unfortunately, this is not possible right now and would require significant rework under the hood. Please put this in the suggestions forum so we can keep track. We wanted to include that in this version, but it was not possible with the time and resources we had.

5.

On 4/6/2024 at 7:17 PM, JW1952TX said:

 

 

In the SMU/Wisconsin game with 1 second left and SMU on the Wisconsin 7 yard line, SMU throws an incomplete pass which ends regulation time and the game goes into overtime.  I think the logic should be to go for the win by kicking the FG as the game was tied 17-17 at the time.  This is #192

I don't know why, but I cannot see you image. Regardless, based on your description of what happened, I can say that this is a known issue, but is the result of some very deep bugs and logic that requires lots of time to debug and test. Given the priorities that we set out for ourselves, fixes for this will likely be in the next major release (the offseason after the 2024 season). We can blame poor clock management by the coach for now (the bane of many football fans).

6. 

On 4/6/2024 at 7:37 PM, subsequent said:

Some more data on kickoff issues. Seems the sim is not sure where to place the ball on some kickoffs (probably because the field flips?). This is resulting in incorrect LOS for the remainder of the drive, and possibly compounding to future drives due to where punts and turnovers take place.

Same as point 1.

7.

On 4/7/2024 at 10:52 AM, Vivid said:

This may have been brought up but the play by play is showing plays that I have listed 0 on in my gameplan. 

Your gameplan is your *ideal* settings, and it is used on the vast majority of plays. However, there are situational gameplans that will override your gameplan in certain situations. For example, if you are a very run-heavy team, your gameplan will not be great for 3rd and long situations where passing is much more desirable. In that case, the engine will substitute a gameplan with default settings for your scheme in that situtation. Thus, there may be cases where plays you normally wouldn't call get used.

Also, this could be the result of option plays if you are looking at the "Point of Attack" column.

8.

On 4/7/2024 at 2:08 PM, alexfall862 said:

Some inconsistency with touches here. FB Michael Carter is present on 40% of formations called up. It should result in about 8 or so touches given the logic as I understand it.

Formation% with FB in it * Play% with Trad Run * Running Distro % * Offensive Snaps = generalized expected carries for a player
Quick Maffs: 0.4 * 0.4 * 0.45 * 77 = 5.54

Since there's only 1 other player on the field with any weight in the FB formations (RB1) my math says Carter should be getting 7% of all plays converted into snaps.

Now, this may be something with starters being rested, but based on that math, Carter should be getting 4.5-5.5ish carries (or 2-4 if he sits the 4th). Yes it could be noise in the rng as far as what is picked, but given past issues manifesting in a similar manner regarding WR3, WR4, and WR5 in past versions of the engine it seems more likely than not is probably a similar issue.

The other thing I've considered is whether having 0 in weight as a RB and 4 as FB may be causing issues where it's returning a 0 when it should return a 4 somewhere?

This next week I'm setting Carter's weight to 10 at FB to lean into that. If not resolved by week 3 I'll removed him from RB carries entirely as well to see if that might be an issue.

(The QBs being 0 and getting touches makes sense because they ran some RPO.)






 

This was the result of having multiple instances of the same player in the same distribution list. The way that the code worked, it would take the first instance of that player on the list and that was it. So in this case, it would refuse to give FB Michael Carter the ball since you had zero weight for him in that section. Interestingly, this is also true for the receiver targets.

So fundamentally, this issue is the result of having one player on a distribution list multiple times (something that is common if you like to play your TE1 at WR or your RB2 at WR), but having different weights for them. It seems like some users are trying to use different weights in those different positions to say "I want player A to have X weight at position 1, and Y weight on position 2". Unfortunately, I think this partially stems from coming from a system where we used to assign weighting based on the position on the field, but now we have transitioned to a system where we assign weighting based on players on the field. Another way of thinking about the new system is: if this player is on the field at all, how much weight do you want them to have for a carry or a target? So, in this example, the engine has two numbers for FB Michael Carter and doesn't know which one is the *correct* one, and it just picks the first one it comes across.

The fix for this is two-fold: 1) Clarification on what weighting is and how to use it, and 2) Creating a standard way to ensure that if there is a discrepancy, we can deterministically determine the weighting to be used in order to protect the engine a bit. To that end:

1) If you have multiple instances of the same player in a distribution list (TE1 player WR4, for example), then it is up to *you* to make sure they have the same weight. DO NOT PUT DIFFERENT WEIGHTS ON THE SAME PLAYER IN DIFFERENT SPOTS

2) In the event that the sim catches such an instance, it will choose the *lowest* of all of the weights assigned to a player. So if you have player A in three different positions on a distribution list, if one of those weights is zero, that player is not getting any carries/targets (depending on what list it is). JUST DOUBLE CHECK YOUR GAMEPLAN. It only takes a few seconds to make sure.

NOTE: This only applies to cases where there are multiple instances of a single player *on the same list*. So above, FB Michael Carter was listed multiple times in the runner distribution. This has zero effect on his distributions on the target distribution list for catching passes.

9.
 

On 4/7/2024 at 2:29 PM, alexfall862 said:

https://discord.com/channels/586207409681334522/1121552692871966760/1226599113886400524



Dropping this here, engine has it correctly ID'd as a play with a fumble, but discord text does not display that it was a fumble.

This is an issue with the discord bot. Tuscan is working on it, but it is low priority as the source of truth for games will be the interface, not discord. The discord bot is a *bonus* type feature and we need to prioritize other things first.

10.

On 4/7/2024 at 2:41 PM, alexfall862 said:

Looking further into the FB carries flagged a similar issue with RBs. 

I had Taylor and Khan on a 5/2 split for Trad Run RB touches. 

At first this seems plausibly accurate given their end of game touches, but 11 of Khan's 16 touches were actually in garbage time. 

Meaning that Taylor took 27 carries in the first 3 quarters to Khan's 5. That's closer to a 6:1 ratio than 5:2, or 85% Taylor compared to the idealized 70%. I have primary RB set to 100% because I don't know what that's supposed to do now with the new carry distribution feature.

Could it be that with Primary RB at 100% other carriers will not get RB touches unless the primary RB subs out due to fatigue? 

In effect, this would mean that the carry distribution for within the RB room is not effective and is actually dictated by the primary RB value in the special teams section.

But this would also raise further questions about FB carries as my backup FB and backup RB (Khan) should have a 1:2 split on carries if they're on the field and thus would have expected 1-2 carries in the 4th to go to my FB as he should have gotten 33% of 40% (he'd be on the field on 2/5th of formations all equally weighted) of my 4th quarter carries.

This should mostly be solved by the fix for point 8.

11.

On 4/7/2024 at 5:27 PM, Spoof said:

Looks like the Safety Score logic isn't set quite right (or Punting/FG Logic is at fault)

 

 

I shouldn't be punting from there, and it called a blocked Punt that went -26 yards a Safety (would have been in GAST 40 range really) and instead of giving GAST 2 points it gave me 2 points and then Georgia State kicked off when they should have received the ball.

Partially I think this is the result of the interface display punts and kicks incorrectly (which should be fixed by now I believe). I have tried to reproduce this and I simply cannot.

12.

On 4/7/2024 at 8:28 PM, tsweezy said:

Something seems odd with the RTP ratios for defenses, but I haven't dug into that many games to see how widespread it was. I know Tulsa got boned super hard by this same thing, the evidence below in the NFL game, and then in my game vs. WVU he got way more "correct" defensive play calls than expected... but also input his RTP numbers reversed for defense. Anyways:

So I tallied up all the offensive plays run out of each formation. So for example Washington ran 8 run plays from Big Pistol. Cleveland had their RTP ratio at 55 against that formation, but on those 8 plays all 8 were called pass defenses. You'd "expect" 4.4 "correct" calls out of those 8 run plays (i.e. they call 4.4 run defenses when given 8 of those looks, to match the 55% run defense they input in the gameplan). Calling 0 of them seems extremely unlikely. As in, (.45^8) unlikely, which is .16% (pretty outside of chance, but technically possible). That's just one example though, as you can see, across all formations and plays the defense consistently guessed "wrong" way more often than you'd expect from just RNG rolls. Out of 171 plays run in this game, there was an expectation that 63 of them would be "correct" defensive play calls. Instead only 17 were. I'm not going to delve into the statistics but just this one game seems like a 1 in 100 million+ likelihood result. When paired with the other 2 games I've seen, it makes me suspect a bug or logic flaw somewhere. (actually, I just told chatGPT to do the P-tests for me, see appendix below):

 

I assumed that the offense would pick a formation, then the defense would "see" that formation, and a RNG would roll a number. If the number was above the RTP ratio set by the defense for that formation, the defense would play a pass defense, otherwise play run defense. However, these results seem to belie that, or indicate that we just saw a vanishingly tiny probability of a game. My only guess is that the formation is called, the pass/run defense is called, and then the offense decides to pass or run with information from the defense (picking which one the defense isn't playing). I can't really think of any other hypothesis to explain it.

Like I said, I only have a very small sample of games so I can't tell how widespread this issue is, but I think worth running a test season or something to check (I can't write anything myself as I have no idea what each team's inputs are for RTP defensive selections)

 

Appendix:

The binomial test results yield a series of p-values for each formation's decision to run or pass. Here are the p-values:
 

WAS	Pistol	Run
WAS	Spread Pistol	Run
WAS	Big Pistol	Run
WAS	Heavy Power Pistol	Run
WAS	Power Pistol	Run
CLE	Near/Far	Run
CLE	Singleback	Run
CLE	Spread	Run
WAS	Pistol	Pass
WAS	Spread Pistol	Pass
WAS	Big Pistol	Pass
WAS	Heavy Power Pistol	Pass
WAS	Power Pistol	Pass
CLE	Near/Far	Pass
CLE	Singleback	Pass
CLE	Spread	Pass

0.0

0.0

0.0

1.34×10−71.34×10−7

0.0

2.87×10−62.87×10−6

4.51×10−44.51×10−4

0.0

0.00437

0.0

0.65671

0.24504

0.29155

0.00968

0.00101

1.0

A p-value close to zero suggests that the actual number of correct calls is statistically significantly different from what would be expected based on the defensive gameplan run percent, assuming that the defensive plays are called at random according to those percentages. A p-value of 1.0 suggests there's no statistical difference between the actual and expected.

For many of these formations, the p-values are very close to zero, indicating that the actual number of correct calls is significantly lower than expected, which could be due to something other than chance, such as a flaw in the simulator's logic or an issue with the gameplan's execution.

This should be fixed by the fix that came from the following issue (13). 

13.

On 4/8/2024 at 12:25 AM, ATonOfLaw said:

GT had a weird amount of runs in the Week 2 Spring Game against Tulane. I've attached the gameplay I had below. I had my pass set to 80%, traditional run set to 5%, and RPO set to 15%.

I counted the play-by-play by hand, so I may have slight human error here, but here's the break down of the actual game play.

I called a pass play 28.1% of the time. I called a traditional run 59.6% of the time. I ran RPO 12.3% of the time.

I don't think I did anything with my gameplay. I had others take a look to see if I had messed anything up. I have no idea why my team decided to run so often. I'm not sure if this is an actual issue or not, but I think something may be wrong.

In week 1 of Spring, I also had more rushes than passes even though I had the gameplan set to heavily favor passing. Thanks for taking a look at this!

So this one confused the hell out of me for the longest time. It took me about a week to figure it out, and it pissed me off because it was so dumb and simple.

It turns out RPO reads were happening on every single play (even if the play called wasn't an RPO play) 

So the fix was simple. Only run RPO reads on RPO plays. Duh (why didn't I have it doing that from the beginning? ).

And really, I wouldn't have been able to figure it out if it wasn't for the fact that Georgia Tech runs basically nothing but passes (thanks @ATonOfLaw). Because schemes that are well mixed don't have any tells like this. It is hard to determine what is a mistake and what is just chance. But when you say "pass a shit ton" and you don't pass a shit ton, that means there is a problem.

So why did RPO reads result in basically the opposite distribution? Because the opposing defense was trying to stop the pass, so the QB was making the read that he should run the ball most of the time since that was where the defense was weak. This also should fix the issue that @tsweezy pointed out in issue #12 above. Why? Because the defense was only "incorrect" because the QB read *made* them incorrect. That is why they were wrong way more often then they should have been.

14.

On 4/10/2024 at 12:53 AM, subsequent said:

 

2pt conversion that is scored on a rush is counting as a rushing TD for the player who scores.

Fixed: 2 point conversions no longer count as TDs for stat purposes on the engine-side.

15.

On 4/11/2024 at 5:12 PM, Spoof said:

 

Another example of the Special Teams/Punt logic awarding an erroneous 2 points, and then play after not making much sense

I cannot seem to reproduce this, as stated earlier. I'll continue to keep an eye on it.

16.

On 4/11/2024 at 7:26 PM, smackemz said:

Backups were in the game, but I don't really ever want a QB throwing to my kicker on 3rd down.  

 

 

Kickers and Punters will no longer be considered valid players to backup any position other than specialists.

17.

On 4/11/2024 at 8:32 PM, Sarge said:

55 13 14 2 11:05 Carolina Panthers 0 0 3 Denver Broncos XP N/A N/A N/A N/A N/A 0 Man Man Man 70385 0 0 0 40 K Michael Rogers's extra point attempt is good.   

56 14 14 2 11:05 Carolina Panthers 0 0 35 Carolina Panthers Kickoff N/A N/A N/A N/A N/A 0 Man Man Man 70385 69984 0 0 84 K Michael Rogers kicks for 64 yards. Fielded at the 1 yardline by RB Hue Khan. RB Hue Khan returns the ball 16 yards. TOUCHDOWN!   

57 14 20 2 10:57 Denver Broncos 0 0 3 Carolina Panthers XP N/A N/A N/A N/A N/A 0 Man Man Man 4999 0 0 0 40 K Elih Obada's extra point attempt is good.   

58 14 21 2 10:57 Denver Broncos 0 0 35 Denver Broncos Kickoff N/A N/A N/A N/A N/A 0 Man Man Man 4999 62531 0 0 86 K Elih Obada kicks for 66 yards. Fielded deep in the endzone by WR Jose Maldonado. WR Jose Maldonado returns the ball 14 yards.   

 

 

Denver got a touchdown for a 16 yard return that started at the 1 in the 2nd Quarter.

Same as point 1.

18.

On 4/13/2024 at 3:41 PM, kgreene829 said:

In the 3 spring games Haloti Thomas has 290 yards and 14 catches on 17 targets and I feel like he shouldn't be that good. He is played at RB but this is him at WR to show that compared to WRs he should suck at catching and route running.

This is partially due to a bug on the interface where we didn't cap the amount of weight a user could put on certain combinations of plays (like Long passes, which is 100% what happened here). Also, dude looks super fast and is probably just blowing by dudes on long bombs. It's easy to catch the ball when no one is around you. Remember long passes don't really care about route running and really only care about speed, which this guy has in spades. Really, I don't think this is an issue, looks like you found a great speed WR.

19. 

On 4/13/2024 at 9:25 PM, alexfall862 said:

Richaun Alsadek is credited with 3 rushing TDs. The play by play data shows no fumbles, and no ints that lead to defensive scores. 

All of UGAs points are accounted for by offensive plays. Somewhere here Alsadek is getting either double credit for a score or a TD is being improperly counted.

EDIT:  This might be the same issue reported here.

M

It is the same as the 2 point issue as you surmised.

20.

On 4/14/2024 at 12:07 PM, alexfall862 said:

Something funky is going on in passing game defense, and ran 100% against the pass this week to try and help ID what the issue was. 

If I'm at 100 pass, I would expect to run pass defense every play of the game. Instead, Broncos ran run defense 80% of the time. 



That's neither 100 or 0, so I don't know exactly what's causing this issue, but it is absolutely destroying any ability to have control over defensive gameplans or look into tuning since all the defensive calls are bunk.

 

EDIT: I may have had the input backwards and 100 is run and 0 is pass, but either way an 80/20 split is not reflecting the desired input.

This is the result of 2 problems.

1) User error. Like your edit said, 100 is all run and 0 is all pass.

2) Situational gameplans. Like I stated in point 7, your defense has some situational settings based on your scheme. So even though you had it set to 100% run, that will only apply in "normal" situations. It seems like your situational defense (correctly) determined that defending the pass was a better option based in the situational gameplan and went with that. Therefore, an 80-20 split makes sense, and is fairly desirable since you can't change your gameplan *during* the game itself, and sometimes what you have in your gameplan isn't a good option for that situation.

21. 

On 4/14/2024 at 12:10 PM, alexfall862 said:

Sacks are reflected in the box score, but not in the play by play.

Aside from needing to be fixed for legibility, this is a severe impediment to the scouting tool as there's not a good way to ID which plays were pass attempts and which were sacks,





It would appear they're showing up in this way: 

This is due to another silly error. The variable in the engine for tracking if a sack occurred on a play was named something *ever so slightly* different than the variable that was on the play-by-play database. Thus, it was never recording sacks on the play-by-play. Players were still correctly credited with sack stats, however.

22.

On 4/14/2024 at 12:16 PM, alexfall862 said:

 

@Sarge including screenshot to help legibility of issue.

Ideas:
- Seems like maybe Khan caught it at the 1, fumbled on the 16, and it was a scoop and score for the Panthers but didn't get the defensive score captured correctly?
- Khan actually returned it 99, not 16 yards?
- IDK?!?!

@

Same as point 1

23.

On 4/14/2024 at 12:24 PM, alexfall862 said:

I set Michael Carter's FB carries to 10, 

We ran 10 run plays from formations with a FB in them at he got 0 carries despite having a 2:1 advantage on distribution compared to the halfbacks, who split carries evenly across the game,

Not sure if this is an issue with Cater being listed as RB3 and getting a 0 on weights there (is it overwriting the FB position weighting for him?). Ideally I'd like to get him into a 20% RB carries and 100% FB carries situation, with around 15 carries a game, but the inputs seem completely borked atm. 

Same as point 8.

24.

On 4/14/2024 at 1:45 PM, tsweezy said:

EDIT: upon further examination 9 of the 11 sacks happened after our starting LG was injured in the 3rd quarter. We had the backup set as the starting RG so I assume the sim just picked someone horrible like the punter to play instead (similar to how DTs are returning punts in blowouts). I guess I'm still curious who got selected but that would make a lot more sense than any of what I posited below (unless anyone else has an issue which I'll keep it up for)

-------------------------------------------------------------------------------------------

Speaking of sacks, can I get a refresher on how those are calculated OR a ballpark of how badly run blocking OL are hurt being in a pass blocking scheme?

These 3 defensive players combined for 9.5 sacks in our game

LE David Eck - 75 speed, 76 strength, 60 pass rush
RE Philip Wilcox - 83 speed, 84 strength, 88 pass rush
DT Lance Nguyen - 45 speed, 65 strength, 67 pass rush

I can see Wilcox having big numbers, but Eck and Nguyen should not be as good at getting to the quarterback as Wilcox, particularly when they're going against on OL of:

92 strength, 89 pass block
90 strength, 84 pass block
86 strength, 79 pass block
94 strength, 88 pass block
85 strength, 81 pass block.

Again, Nguyen and Eck having 6 sacks in 40 plays against that line seems crazy high to me. My best guess is that this was a hot fix to try and make long passes hurt more, but I'd like to know if that was the case or if there was just some crazy set up stacked maluses somehow, or if Morrison is just uniquely the worst player in the sim at not getting sacked.

The problem is, right now we don't know who got substituted as we don't know who is on the field each play (something we are planning to include somehow in the future). Whoever it was it wasn't very good, but that doesn't mean it was your punter. It could have been anyone with poor pass blocking, even a lineman.

25.

On 4/14/2024 at 4:53 PM, Bundy said:

61	7	0	2	Minnesota Vikings	07:03		35 Minnesota Vikings	Kickoff	N/A	N/A	N/A	N/A	N/A	0	Man	Man	Man	85	K Mike Tsao kicks for 65 yards. Fielded at the goalline by RB Fish Liggins. RB Fish Liggins returns the ball 15 yards.
62	7	0	2	New England Patriots	06:55	1st and 10	20 New England Patriots	Pass	Quick	Double Tight	4-3 Heavy	Quick	Pass Defense	0	Man	Man	Man	0	QB Rashaan Lyons throws it... and it's incomplete. Pass intended for RB Jerry Lawson.

 

Not sure if the "fielded at the X" is just flavor but if its not, then wondering why a KR fielded at the goal line and returned 15 yards came out to the 20 (or is this an issue of the return yards being wildly off and it wasnt actually returned out of the end zone)

Same as point 1.

26.

On 4/14/2024 at 5:41 PM, Bundy said:

85	7	0	3	Minnesota Vikings	12:22	2nd and 4	18 New England Patriots	Pass	Short	Singleback Gun	4-3 Base	Short	Run Defense	0	Man	Zone	Zone	3	QB Christian Sinnock throws to WR Adrian Daniels complete for 3 yards. Tackled by SS Johnny Nguyen. PENALTY: Offensive Holding. Offense. Player: OT Herman Knott. 0 yards.
86	7	0	3	Minnesota Vikings	12:05	2nd and 4	18 New England Patriots	Pass	Short	Spread	4-3 Base	Short	Run Defense	0	Man	Zone	Zone	5	PENALTY: Illegal Shift. Offense. Player: WR Alex Williams III. -5 yards.

 

Im sure this has come up before but logging it for reference, Holding didn't subtract yards, was just a replay of downs/ate some clock

 

EDIT:

but this other holding call did work

100	7	0	3	Minnesota Vikings	06:39	2nd and 5	20 New England Patriots	Run	Inside Run Left	Singleback Gun	4-3 Base	Inside Left	Pass Defense	1	Man	Zone	Zone	14	RB Yogi Oladipo carries for 14 yards. Tackled by ILB Ariel Calderon. PENALTY: Offensive Holding. Offense. Player: OG Mark Killough. -10 yards.
101	7	0	3	Minnesota Vikings	06:19	2nd and 15	30 New England Patriots	Pass	Play Action Short	Singleback	4-3 Base	Short	Pass Defense	1	Man	Zone	Zone	4	QB Christian Sinnock throws to WR Alex Williams III complete for 4 yards. Tackled by OLB John Damiano.

Unfortunately this is another one of those "not enough time and hard to fix" issues that I cannot commit to fixing before the regular season starts. The problem with penalties like this is there are so many tiny edge cases that cause problems. So for now we will have to chalk it up to bad refs. Break out the blind ref gifs.

27.

On 4/15/2024 at 4:29 PM, alexfall862 said:

Play 161 had both a kickoff and a 2pt conversion in the same play? Is this related to the issue with the Hue Khan 16 yd kickoff TD and the successful 2pt conversions counting at TDs?

This is play 161 of the Week 2 Rams/Bucs game.

I could not reproduce this. I believe it was interface-side and was fixed.

28.

12 hours ago, alexfall862 said:

Dunno if the other suggestion threads are supposed to be active - but I'd kill for a 4CB dime package in addition to my backup SS.

Base package has 4-3 and 4-3 over and it's a nice way to change things up, would be cool to have that for dime. If wrong channel, that's cool, just trying to drop it somewhere before I forget.

For everyone, please put suggestions in the respective suggestion forum for CFB (or NFL if you are in the NFL and it is NFL-specific).

[Bingo415]

Thank you for taking the time to respond. Dev team is great.

[Wahoo]

Not sure if there is a new thread to post in but:

 

Looks like penalty messed with the yardage counter. FAMU-FSU Week 1

 

[smackemz]

In the same spirit that how many long passes can be thrown was limited for realism - it probably makes sense to  not allow teams to pass 100% of the time or run 100% of the time.  

Even if the percentage is small (10 or something) - it would make for more realism.  Teams simply do not only pass or only run. 

[tsweezy]

30 minutes ago, smackemz said:

In the same spirit that how many long passes can be thrown was limited for realism - it probably makes sense to  not allow teams to pass 100% of the time or run 100% of the time.  

Even if the percentage is small (10 or something) - it would make for more realism.  Teams simply do not only pass or only run. 

Related to this:
I think the fix is just a change in how schemes validity is determined. Right now the pass/run/RPO % is just horizontally summing the distribution across formations. But it *seems* like the engine chooses formations based on weights, and then chooses a play based on that formation. So if you have a formation with a very high running distribution and very low weight: i.e. 

This will show up as your overall gameplan being 50% run even though you only *actually* run the ball 1% of the time. I think the calculation for gamplan %

needs to just be recalculated as (sum across formations of) formation weight/100 * play distribution / total distribution. Then the reported Percentages will be accurate to what your gameplan is, and the regular enforcement of percentages by scheme should work