Gustav

Week 3

I write goalers in basically every article for this reason and I didn’t know it changed. Whoever came after my goalers, we need to talk.

I'm testing the made up rating system I made in my last article by making my VHL bracket solely based on it. Go Malmo I guess? Unless the Bears make it all the way too.

And somehow, because I SOMEHOW forgot to mention the winners... CONGRATULATIONS TO OUR WINNERS, THE MAFIA AND THE PIRATE!!!

By the time these threads get to the point where we're killing a confirmed good guy, they're usually pretty quiet. This was no different. Usually, Town being wrecked happens a little bit later than Night 1. Unknowingly to Town, @Doomsday was the JAILOR who did, in fact, jail @Ricer13 on that first night. By chance, the MAFIA came for Doom immediately. On Night 2, Town lost three of its members. After the most important Town role fell on N1, the second-most important--MAYOR @LucyXpher--was taken out by @NSG the JUGGERNAUT. With half of Town out immediately, including its two largest fighting chances of winning and both Protective roles, it was essentially all we could do to watch our thread turn quiet and uneventful while evils sorted themselves out. We conclude our game with a fairly expected, but still allowed for, choice to lynch @jhatty8, the Town's last remaining citizen. Given that there is no more Salem left to Town, I think I get to acknowledge Hatty's refusal to adopt the new role names--he was an ESCORT. However quiet this game may have been, I sincerely hope that you had fun playing it, and that you learned a lot and would come back if it was your first time trying. As always, thank you so much for signing up! As long as I can bring you a fun time, I will never consider it work. @Adrest245 - RETRIBUTIONIST @Advantage - PIRATE @Da Trifecta - PSYCHIC @Doomsday - JAILOR @eaglesfan036 - VAMPIRE @GoodLeftUndone - MAFIOSO @jhatty8 - TAVERN KEEPER @Lemorse7 - DOCTOR @LucyXpher - MAYOR @N0HBDY - CRUSADER @NSG - JUGGERNAUT @omgitshim - GODFATHER @Ricer13 - AMBUSHER @SleepyPlayz - FORGER @Spartan - SPY --> VAMPIRE

We are off to another day--whether to call it is still not a decision I can make confidently. @Adrest245 was attacked by the MAFIA last night. He was the RETRIBUTIONIST. Have fun choosing your own destiny today! NSG's last will from yesterday that I missed: List of living players (4) @Advantage @GoodLeftUndone @jhatty8 @SleepyPlayz

Gustav is watching football, and that was admittedly enough time for the vote to swing. Gustav is also on his phone. NSG was a JUGGERNAUT. Sorry. List of living players (5) @Adrest245 @Advantage @GoodLeftUndone @jhatty8 @SleepyPlayz

^Eagles was killed by a PIRATE and I'm dumb for not mentioning that.

Gustav wakes up with a sore shoulder and a lack of desire to go to work. Some of Salem wakes up with him. Last night, @Da Trifecta was killed by the MAFIA. He was a PSYCHIC. @Ricer13 was taken out, this time by a CRUSADER. He was the AMBUSHER. Finally, @eaglesfan036 has made it up to this point by first claiming to be a SPY and then going quiet. Eagles was... The night phase will come when it always does. List of living players (6) @Adrest245 @Advantage @GoodLeftUndone @jhatty8 @NSG @SleepyPlayz

Pictured: me counting the day's posts. Such is life when Town gets hit hard and early, I suppose. We head into N4, and who knows what's coming? Gustav will have the next day phase open at 9:something, probably. List of living players (9) @Adrest245 @Advantage @Da Trifecta @eaglesfan036 @GoodLeftUndone @jhatty8 @NSG @Ricer13 @SleepyPlayz

2,075 words; week 2

It's been a long time since I wrote a numbers-based article. Of course, you're all familiar with all that time I spent talking about myself. Chances are that you're also familiar with things like my Hall of Not Bad series, which I really do promise hasn't died yet. But it's been a while since I wrote a standalone article on a new topic, and I'm finally back to give you one of those. I've always been curious about how hybrid attributes affect player builds. I'm still much better at looking at player builds through the lens of their STHS attribute equivalents, and I also generally try to stay mindful of the idea that hammering away at a hybrid attribute might only get you a point or two in the grand scheme of things. This was something that frustrated me a little bit when we drew up the hybrid system. Especially as attributes get more costly, a system that was already meant to make things more costly will end up making that effect even stronger. So, as I build, I should aim to build in as efficient a manner as possible. Indeed, I've always referred to this concept as "TPE efficiency", but I never ended up doing more than casually thinking about it and I've never attempted to run any numbers on it. Ideally, one would expect a good build to do a collection of things: Be the result of good earning (i.e., have a lot of TPE); Be as efficient as possible (i.e. upgrading real STHS attributes as much as possible without losing much to the hybrid system); Conform to a "meta"-type build (have strong Scoring and Defense without significant upgrades to Passing). So, I think it would be worth our time to take a look at a few things. Which players have reached a high TPA level, while not being held up much by STHS, and while building well--doing both of those other things while going for builds that target hybrid attributes that aren't often thought of as efficient adds? Let's find out. 1. TPE Efficiency I'm going to define TPE Efficiency as (STHS TPE)/(Hybrid TPE). If a player somehow spends 500 TPE on a build that is also 500 TPE under the old update scale, that player's TPE Efficiency will be equal to 1. The new system is just about universally more restrictive than the old one, though, so everyone will end up with a decimal value. On the surface level, TPE Efficiency is best of as high as possible. Across the league, the spread looks like this: And I'll be honest, this actually really surprised me. My perception was that as TPA gets higher and as the update scale starts to take a stronger hold, STHS upgrades will become harder and harder to come by, and builds will become less efficient. It turns out that I was wrong about this--of course, the old update scale also got harder as it went on. Ours actually seems like it's making things a little bit easier than it used to be for high-earning players than it used to be--granted, I don't think this is a huge concern because it's not all that strong of a correlation, but it's definitely there. For instance, lots of players above 600 TPA manage builds that are over 60% efficient, but none below that point. Builds below 50% efficiency are also a rare occurrence above 700 TPA or so, but there are lots of them below that point. It actually may be the case that there's a peak just above 800 TPA, at which point our update scale starts to become more difficult again, but I don't have enough data points that I think I can say this confidently. I've noticed that aside from TPE, most of the most efficient players are forwards, and within that, most of those are centers. I'm not quite sure what's causing the discrepancy between forward and defense in general, but a center would benefit from the need to add to Faceoffs, that being one more attribute that would be upgraded at low cost. In general, these differences aren't huge. If you're curious, the league's most efficient defenseman is @Daniel Janser's C.G.Y. Janser, weighing in at 61.8% efficient. All in all, once you're above 800 TPE, most players in the league are between 50 and 70% efficiency. A couple players of note in this category are: Most Efficient: 1. Miner vHLer | | LW | @Rhynex Entertainment | 847 TPA | 77.8% Efficiency 2. Viktor Jensen | | C | @Advantage | 850 TPA | 74.1% Efficiency 3. Shubham Anand | | RW | @Will3 | 847 TPA | 71.8% Efficiency Least Efficient: 1. Chris Reynolds Jr | | LW | @Crstats23 | 611 TPA | 38.0% Efficiency 2. Sunset Moth | | D | @McWolf | 551 TPA | 38.7% Efficiency 3. Chris Sanzone | | LW | @handfullofbubba | 642 TPA | 38.9% Efficiency Honorable Mentions: -Mina (@Baozi) is the only player above 800 TPA and below 50% efficient. Mina's 1085 TPA translate to 524 TPA in STHS--a good deal under Miner vHLer's 659 despite being over 200 TPA ahead on the portal. -Calgary's Raimo Tuominen (@SlapshotWrangler) is building impressively, 4th in overall efficiency despite being above 1000 TPA. Tuominen is the only player above 900 TPA to be over 70% efficient. -Davos' Joseph Dubois (@Dubois) is inactive and would be off to an awesome start if we were only earlier on in his career. Dubois is that little peak right around 400 TPA, currently rocking a 53.5% mark. That's very impressive for the TPA level! Efficiency can tell us a lot, and even though I knew there were often big differences in those STHS attributes that weren't always evident in the hybrid totals, I wasn't aware that we had differences as big as we do or even that TPA can influence how much efficiency is even possible at a given level. It does not, however, tell us the whole story, and that's where we need to try to quantify builds. 2. The Meta Index So, I'm going to introduce a new number called the Meta Index (MI), which I'll define as MI = ((SC/PA)-1)*(SC/99)*(DF/99). The first part of the formula, which I've put in red, weighs Scoring and Passing against each other ratio-wise. Equal Scoring and Passing will result in an MI of zero, while values of Scoring greater than Passing will make MI positive and values of Passing greater than Scoring will make MI negative. This value is then multiplied by the blue part, which is a measure of how much Scoring and Defense have been upgraded. This results in a generally higher MI for players with higher TPE, which I don't personally mind because those players are both more valuable to their teams and more locked in to their builds (a low-TPE player can become meta, but a high-TPE player either is or isn't unless they reroll). So, let's take a look at player builds as they conform to the meta. Of course, a number that I drew up to get larger with TPA ends up getting larger with TPA. There's no surprise there. What I do think is interesting to see is that you can certainly tell which players have locked themselves into the meta build strategy and which have not. Block out every MI below .100 and you can draw a very clean trend line through those top points--which, to me, indicates that there's a pretty constant rate of increase in MI as those builds progress and the meta doesn't really become any easier or harder to pursue with time. Of course, I can always list out the following: Most Meta: 1. Harry Callahan | | RW | @dstevensonjr | 1330 TPA | MI .279 2. Gianfranco del Rocco | | LW | @Victor | 1051 TPA | MI .276 3. Three players tied with MI = .267 Least Meta: 1. Felicia Hardy | | D | @JCarson | 905 TPA | MI -.175 2. Lionel Collberg | | D | @Masu Chan | 830 TPA | MI -.147 3. Tord Yvel | | D | @Trazan | 618 TPA | MI -.140 It's true that our top players are forwards, and our bottom three all play defense, but I don't think there was as strong of a difference position-wise as there was before. I'd explain the difference here as being up to differences in the ways people choose to build voluntarily. For instance, I've certainly seen more people playing defense and not upgrading Scoring, just because that's what's seen as making a "defensive defenseman" to some. What about efficiency? Originally, the Board of Gustav chose to make some of the ratios the ways that they are in an attempt to make pass-first builds more accessible to those who wished to build outside of the meta. Were we to compare efficiency to MI, I would hope that some of those low (and even negative) MI values would correspond to higher efficiency values--because that was the intent, wasn't it? It doesn't look like it upon first glance, and I think it's a fair takeaway that the differences in ratios needed to upgrade Passing and Scoring really aren't influencing overall efficiency too much. However, I would argue that pass-first builds are slightly more efficient, because we're seeing MI increase pretty steadily with TPA, we're seeing higher efficiency at higher TPA, but we're not really seeing high efficiency at high MI. I would argue that low-TPE builds are in fact slightly more efficient--Miner vHLer, being the most efficient player we have, certainly seems to have taken advantage of that to some extent. Whether it's worth going for a pass-first build for the sake of efficiency, though, doesn't seem to be a question with a positive answer, because I'd take that effect as very slight. As a side note, I will always say that you should build however you'd like. If your MI is low, don't worry about conforming unless you'd actually like to. But as it's clear that a player can be efficient, and that a player can conform to the meta, but neither one of those is necessarily true for a given player, it's certainly also clear that TPA isn't going to tell the whole story--and neither is a casual glance at whether a player has the cool stuff upgraded. We need numbers, and I need to invent another arbitrary stat. 3. Build Grade I'm going to (again, arbitrarily) define a new number called Build Grade as BG = MI + ((STHS TPA)*Efficiency/1000). This is another number that should get higher with TPA, but it's more complicated than just that. Here, I'm taking the MI and adding it to that big term on the right, which takes STHS TPA (not portal TPA) and weighs it against efficiency. So, we're creating a number that factors in how desirable a build is with MI, more directly factors in TPA level (but only considers the one that actually matters), and rewards those who have gotten to that point efficiently. It's a rough approximation, but I think it does a pretty decent job of giving an answer that lots of people want: who is the best-built player in the VHL? And of course it gets larger with TPA, but there's still a good deal of variance here that makes this view worth it. As you can tell, for instance, the current leader in TPA is not the VHL's best-built player. Nor is it necessarily a guarantee that you'll end up with one of the best builds out there if you earn a lot--perhaps those around 1200 who may be underperforming could compare their stats to that big group just over 700 and get back to me about what you see. In all, though, please take a look at the final results and let me know what you think: I've also put the average build grade per team on a chart, which agrees OK enough with the standings in most cases. It's not perfect, and that's the beauty of statistics: But yeah, there you have it--my first standalone numbers-based article in a long time! I make no claim that my own numbers are anything beyond "I just felt like it," but I'm pretty happy with the result and I learned some cool stuff along the way. Hopefully you get as much out of reading this as I did writing it! If you'd like to see where you fall on the list, or if you'd like to play around with the numbers on your own, you can find my spreadsheet here.

The sun once again rises on Salem, and the town is once again the slightest bit smaller. @omgitshim was killed last night, the second citizen to die to a JUGGERNAUT so far. He was... Additionally, @Spartan was finally taken out by the MAFIA after spouting off some bullshit about "Windiana" the other day in the game server. He was... Night phase should--once again--be on time. Let's have a good day! List of living players (9) @Adrest245 @Advantage @Da Trifecta @eaglesfan036 @GoodLeftUndone @jhatty8 @NSG @Ricer13 @SleepyPlayz

Gustav is on his phone tonight. It’s Saturday night and time to party (read: go out anyway after no one wanted to hang out). While Gustav is exploring the depths of the only bar within walking distance, the day phase ends. And no one dies, yet again. Three days, no lynches, and time for another night. Gustav will be sleeping in tomorrow again, and no, he sees nothing wrong with continuing to refer to himself in the third person. List of living players (11) @Adrest245 @Advantage @Da Trifecta @eaglesfan036 @GoodLeftUndone @jhatty8 @NSG @omgitshim @Ricer13 @SleepyPlayz @Spartan

Night 1 can often be uneventful in Salem. Night 2...perhaps not as much. We start off by checking on @N0HBDY, who spent yesterday hinting at the possible existence of a CRUSADER. It's possible that this may have ticked off the MAFIA somewhat, who came by to visit last night. It's indeed true that there was a CRUSADER in this game, and that N0HBDY knew what he was talking about. One town down. Second of all, last night revealed the existence of a PIRATE. @Lemorse7 was found dead this morning, with multiple wounds from a weapon yet to be revealed. He was--again, to the disappointment of the Town--a DOCTOR. And finally, the Town finds that @LucyXpher, a first-timer, has been taken out by a JUGGERNAUT. Poking around a little more, we also learn that Lucy was... Gustav expects to be around at the standard time tonight, but that may change if I end up with any plans. List of living players (11) @Adrest245 @Advantage @Da Trifecta @eaglesfan036 @GoodLeftUndone @jhatty8 @NSG @omgitshim @Ricer13 @SleepyPlayz @Spartan