Gustav's Big Fat CK/DI Manifesto

[Gustav 6,407]

Looks at media spot claims

Sees I have nothing for the coming week(s)

Pounds on this stupid E key which occasionally gives me two Es when I don't want them and makes writing a pain in the neck no matter how much I try to clean it out or fix it

 

All right! Enough of that nonsense; let's get into it.

 

The greatest first-gen resource ever known to the VHL community and the font of all holy knowledge, Gustav's Guide to Building, contains this bit of wisdom:

Quote

high checking and high discipline is just a waste of TPE!

 

...but how true is this, really? Is it actually worth your time to upgrade both? Will taking it to an extreme give you a physical-but-clean player? I'm going to attempt to quantify this as accurately as I possibly can, using data taken from real VHL seasons, with real VHL teams, real VHL players, and real VHL sim settings that don't leave the tester wondering "well, this is all well and good, but is it at all accurate the way the league does it?" (Legal disclaimer: I did not steal a league file, as much as I wish I had one to work with.) Instead, I'm using data from real seasons--S72, to be exact, because I did the research for this article a long time ago and I've just been too lazy to put it in the form of actual words claimable for some sweet hot TPE.

 

First of all, let's take a look at the two attributes in question, and what I've always thought they've done:

Checking: Increases the number of hits a player records. There also tends to be a fair amount of positive correlation between checking, hits, and penalty minutes, making this a bit of a double-edged sword.

Discipline: Decreases the number of penalty minutes a player records. Upgrading discipline also tends to reduce hits.

 

Ideally, a player skilled in both should be able to record relatively high numbers of hits with relatively low numbers of penalty minutes. At least, one would think. I am coming into this article with my own biased hypothesis based on what I believe my own personal experience with building to have been: high Checking and high Discipline does nothing but look cool on your player page. Let's quantify.

 

yes, the title of this should be PIM vs. HIT. I deeply apologize and hope @McWolf doesn't go full nerd on me and beat me to death with a slide rule.

 

Positive correlation, you say? Positive correlation! It's a well-known fact in today's VHL that the more hits one records, the more penalty minutes they'll rack up (in general). As this has nothing to do with attributes, I'm merely using it to establish something nice early on--what effect does Discipline have on this chart? What effect does Checking have on this chart? Let's take a look at that second bit here:

 

 

Does anyone notice that both of these charts look very similar? Of course, they're not the same--it's worth noting that, looking at the equations of these charts, checking seems to increase hits at a greater rate than it increases penalty minutes. But there's still a nice upward trend for both, and the fact that we've got a lot of players with CK at 40, 70, and 80 (see those vertical columns?) provides us a nice and convenient opportunity to extract data from only those groups--which I'll get into later. 

 

 

Read the subtitle here--this chart doesn't do anything; I'm just throwing it out there. In fact, the points were so spread-out that Google Sheets couldn't even draw a trendline for it.

 

 

Same thing goes here. One might be inclined to say that, based on how scattered everything is, Discipline doesn't do anything. Let's not rule that out yet--keep in mind, players with DI at 40 could have CK at 40, or they could have it at 99. If Checking really does bring penalty minutes up, is Discipline really doing anything to bring them down? Enough of looking at the attributes themselves; let's have a look at ratios.

 

 

I'm taking a ratio between Checking and Discipline here because, let's face it, we're talking about Checking and Discipline. Just looking at one or the other isn't really going to show us anything, as evidenced by the disaster that is those Discipline charts. Here's one that makes things look a bit off, though--in theory, if Checking increases penalty minutes and Discipline decreases them, one would expect a chart showing the ratio between those two to spit out data that's more consistent and closer in general to the trendline. The R-squared value for this chart, though, is lower than that of the value of the PIM vs. CK chart, meaning that the numbers it put out correlate less with the trendline. We've again got a big vertical column at 1.00, though--are people on the lower end of that column players with high Checking and high Discipline? Let's take a look at hits:

 

 

At this point, it's worth bringing up something interesting: in this population, no player with equal CK and DI recorded 150 hits, and most fell below 100. As might be expected, there's again a positive trend here--and, as exhibited above with the CK charts, hits increase at a rate greater than penalty minutes. What's interesting here is that, as we move from Checking itself to the relative position of Checking as it relates to Discipline, this difference gets even more pronounced:

 

148/86.2 = 1.72 (slope of HIT vs. CK/DI divided by slope of PIM vs. CK/DI)

3.35/2.11 = 1.59 (slope of HIT vs. CK divided by slope of PIM vs. CK)

 

What this at the very least suggests is that factoring in Discipline actually makes Checking look better--having Checking a good deal above Discipline still gets you a better hits:penalty minutes ratio, and having it a good deal above Discipline seems to actually have this effect more strongly than just having decent Checking and having Discipline wherever.

 

Now for the big one...

 

 

In theory, high Checking and high Discipline should, again, give a player lots of hits, with not a lot of penalties. And a look at this chart is actually fairly encouraging--minus one outlier around the 1.125 mark that I didn't bother to check and whatever the hell Dean Clarke did (44 hits and 4 penalty minutes with 61 CK / 40 DI, a fairly small sample size leading to a bad point on this chart; dammit @Kyle!). Now, in general, it appears that hits are more or less just as clean across the board--but what about that vertical line at a CK/DI ratio of 1? Could it be that upgrading both more or less evenly to a significant extent will give you more hits and cleaner hits?

 

I decided it was time to look into that, and found that out of all the players with equal CK and DI in S72 (37 in total), only four had anything other than 40/40:

 

-Micheal Gary Scott (@Motzaburger) - 90 CK / 90 DI

-Charles Drumm (@frescoelmo) - 75 CK / 75 DI

-Luciano Valentino (@Kekzkrieg) - 75 CK / 75 DI

-Ricky Johnson (@Midnite) - 70 CK / 70 DI

 

There were more players with similar ratios (in fact, I got the idea for this article in the midst of making fun of @Berocka for fairly unimpressive checking numbers with his 78/70), but for the sake of consistency I'm choosing to take a look at those who have both attributes upgraded exactly equally. The data may be found here, and, let me tell you, it's not the greatest news if you've gone and maxed out your player in both:

 

 

The "high" group (that is, Scott, Drumm, Valentino, and Johnson) did manage to put up more hits than the 40/40 gang, but that is the only advantage they've got here. They took more penalties, and their HIT/PIM ratio is much lower--meaning that the hits they gave out were less clean than those given out by players with not a single point in Discipline. This is basically like getting 50 games into the schedule and having a statline looking like the one on the right--and then from that point on to the end of the season, having every other hit you record put you in the penalty box. That's not good.

 

It's also worth noting that I took the average HIT/PIM ratio for all players with a CK/DI ratio below 1 (that is, Discipline is higher than Checking), and got...this.

 

 

That gives you just as many hits as the group at 40/40, with more penalties and a worse ratio. So, I'll conclude with this:

 

TL;DR: Discipline is entirely worthless.

 

I do not see even a single reason why upgrading Discipline in any capacity is beneficial to the growth of a player. If you want to go and build a two-way player, build a two-way player by upgrading Checking. Upgrading Discipline with Checking will dial you back down to a level equal with--if not worse than--where you started. Want to upgrade Discipline only to avoid penalties? Cool, that also does nothing and may even make you slightly worse. 

 

That's enough stats for me; see you in three weeks. Have I hit the word count yet? I think I have...

 

 

Yes! I have--1,552 words. See you in three weeks indeed.

[McWolf 3,115]

did you try to correlate with the gap between CK and DI?

[Gustav 6,407]

Just now, McWolf said:

did you try to correlate with the gap between CK and DI?

 

I did; the charts looked just about identical to the CK/DI ones so I didn't bother mentioning it.

[McWolf 3,115]

Just now, GustavMattias said:

 

I did; the charts looked just about identical to the CK/DI ones so I didn't bother mentioning it.

did you try looking at H/20 or PIM/20 to incorporate the time on ice in the equation?

did you try *more dumb stuff*?

[Gustav 6,407]

Just now, McWolf said:

did you try looking at H/20 or PIM/20 to incorporate the time on ice in the equation?

did you try *more dumb stuff*?

 

smh there's no smh reaction

 

Maybe that will be what I do in three weeks.

[Doomsday 4,130]

I have never spent a single TPE on discipline with any of my players and I clearly was right to do so. 

[Victor 10,940]

12 minutes ago, GustavMattias said:

TL;DR: Discipline is entirely worthless.

Yes

[eaglesfan036 4,603]

Awesome article! Also, kind of dumb that the sim engine has so many useless or downright harmful stats

 

These articles are a great example of why we need to find a better sim engine IMO

[McWolf 3,115]

Was the article a sneaky way of asking me to rerool my discipline?

[eaglesfan036 4,603]

Just now, McWolf said:

Was the article a sneaky way of asking me to rerool my discipline?

a sneaky way of asking you to retire

[McWolf 3,115]

Just now, chatfan036 said:

a sneaky way of asking you to retire

harsh

but fair

[Garsh 1,190]

Good work.

[Beaviss 4,957]

[Fire Tortorella 2,653]

I could have told you discipline is useless in 3 words instead of 1,552.

 

2 hours ago, chatfan036 said:

These articles are a great example of why we need to find a better sim engine IMO

 

Unfortunately, a different sim engine (assuming you mean not STHS at all, not necessarily what version of STHS we're using) would mean we lose a good deal (if not all?) of our automation...I think. I'm not exactly a portal aficionado but I'm almost positive that's the case.

[DMaximus 1,046]

I'd like to see an analysis of PIM and Goals Against and how that fits into all this. It seems like the average PP is around 20%. You could roughly equate that to 10 PIM (5 penalties) = 1 Goal Against. So someone who records 100 more PIM than normal because they have their CK at 80 and DI at 40 is causing their team 10 goals against.

Edited September 8, 2020 by DMaximus

[JeffD 648]

11 minutes ago, DMaximus said:

I'd like to see an analysis of PIM and Goals Against and how that fits into all this. It seems like the average PP is around 20%. You could roughly equate that to 10 PIM (5 penalties) = 1 Goal Against. So someone who records 100 more PIM than normal because they have their CK at 80 and DI at 40 is causing their team 10 goals against.

I was going to post nearly the exact same question. With how good PPs are in the league are high hit players actually detrimental to their team? 

[Beketov 9,025]

14 hours ago, flyersfan1453 said:

Unfortunately, a different sim engine (assuming you mean not STHS at all, not necessarily what version of STHS we're using) would mean we lose a good deal (if not all?) of our automation...I think. I'm not exactly a portal aficionado but I'm almost positive that's the case.

We could potentially rebuild a bit of it but yes, that would be the case. The portal is built around STHS' database as well as it's input and output for player's and stuff. It's very much not an easy switch. Just look at the EFL or SBA. People have said both have better engines than ours (I'd definitely argue the case against the EFL's being better when only like 3 attributes matter to most players) but both of those have a major flaw we don't have: no player inputs. That means going back to good old manually updating players, no thanks. In addition the "new sim" crew always seem ti make the assumption that any different engine must be better than ours, that's not necessarily the case. From hat I've heard with the SHL changeover people don't like the new engine at all. So that sounds like a hell of a lot of trouble with very little to no guarantee of anything better.

 

18 hours ago, GustavMattias said:

TL;DR: Discipline is entirely worthless.

I put 5 into it to see what would happen. I don't think it really did anything but 5 TPE wasn't exactly a hefty price to pay haha

[Fire Tortorella 2,653]

1 hour ago, DMaximus said:

I'd like to see an analysis of PIM and Goals Against and how that fits into all this. It seems like the average PP is around 20%. You could roughly equate that to 10 PIM (5 penalties) = 1 Goal Against. So someone who records 100 more PIM than normal because they have their CK at 80 and DI at 40 is causing their team 10 goals against.

 

1 hour ago, JeffD said:

I was going to post nearly the exact same question. With how good PPs are in the league are high hit players actually detrimental to their team? 

 

Without any stats to back this up, I generally look at hits as a proxy for takeaways. It's not going to be a 1:1 ratio, but I think it's a good estimate.

[frescoelmo 1,276]

Wait what did I do

[Gustav 6,407]

9 hours ago, DMaximus said:

I'd like to see an analysis of PIM and Goals Against and how that fits into all this. It seems like the average PP is around 20%. You could roughly equate that to 10 PIM (5 penalties) = 1 Goal Against. So someone who records 100 more PIM than normal because they have their CK at 80 and DI at 40 is causing their team 10 goals against.

 

9 hours ago, JeffD said:

I was going to post nearly the exact same question. With how good PPs are in the league are high hit players actually detrimental to their team? 

 

This could be good to look at as well; thanks for the article ideas 

[Gustav 6,407]

Week deux

[Gustav 6,407]

Claiming my third and final week of this with this post. Possibly more to come in the somewhat-near future if I've got the time or the motivation.