Friday, 05 January 2018 15:59

THE WISE OLD WEASEL-- EXPLOSIVENESS PART 2

Written by
Rate this item
(0 votes)

The Wise Old Weasel's Five Factors

Factor 1: Explosiveness

Part 2

DOES PREVENTING BIG SCORES MATTER?
In our last installment, we established that players that make Royale have some degree of explosiveness- an ability to have good (40+ point) or great (50+ point) games. While league winners and Royale participants are about equally likely to top, league winners are much more likely to have explosive games.

In football, the sport that inspired this series, there are two aspects to explosiveness: having it on offense and preventing it on defense. Both are critical and positive factors in achieving victory.  This installment evaluates whether this is also true for diplomacy- does having an ability to stop big scores make you more likely to win the league?

The data says that Royale participants and League winners are better at preventing explosiveness. As we’ll discuss more later, figuring out the “why” behind that is a lot more difficult than it is on the “offensive side of the ball.”

NICE SETUP, JAKE. CAN YOU STOP TEASING THE RESULTS AND GET ON WITH IT?
Hey, cool it, bolded alter ego! I get metagaming makes you cranky. Just hit the jump and we’ll get into the practical data.

Hit the jump to see just the role of (explosiveness) defense in winning championships.

SO, IF SOMEONE IS GOOD AT EXPLOSIVENESS DEFENSE, THEY WOULD BE GOOD AT PREVENTING OTHER PEOPLE FROM GOOD AND GREAT SCORES?
Very astute, alter ego. By definition, someone who prevents explosive scores (AKA good Explosiveness Defense) would cause games to be less likely to result in said scores.  To get you grounded, let’s take a look at a typical season of WCW Diplomacy in terms of explosiveness.

"AGGREGATE WEASEL"
# Boards: 34.6
# Good scores (percent): 19.14 (54.5%)
# Great scores (percent): 9.29 (26.4%)

AGGREGATE WEASEL WOULD BE A GREAT BAND NAME.

Sure would. Auditions next week.

OKAY, SO HOW DOES THIS (explosiveness) DEFENSE THING WORK?

So glad you asked. To answer this, we bring back our old friends, Toppy (Projected Royale 1 seed) and Caboosey (Projected Royale 7 seed) Weasel. We calculated Explosiveness D by calculating the number of good or great games that were had by opponents that Toppy or Caboosey played on.

"Toppy Weasel"
# Boards: 19.38
#Good Scores Allowed (percent): 5.63 (30.2%)
# Great Scores Allowed (percent): 2.38 (21.2%)

 

"Caboosey Weasel"
# Boards: 10.42
#Good Scores Allowed (percent): 1.86 (17.76%)
# Great Scores Allowed (percent): 3.43 (32.17%)

I SEE THAT THOSE ARE NUMBERS. I DO NOT SEE WHAT THEY MEAN.

Oh, stop. You’re too hard on yourself, alter-ego. It’s apparent that the other 6 players on Toppy or Caboosey’s board are less likely to explode (relative to the rest of the season). They’re about equally likely to surrender an explosive game (51% vs 50%), but Toppy’s won’t let you have a great game.

CAN’T THAT BE EXPLAINED AWAY BY OTHER FACTORS?

Honestly, yes. Diplomacy is a seven player, zero sum game that is very different from football. One player winning or being in the hunt or even not dying makes it less likely for any other player to break out. When you factor in that Five Factors evaluates not just games, but seasons… well, it is important to keep any defense stat in context with other defense stats.

DAMMIT KID ANSWER THE QUESTION- DOES IT MATTER?

Yes, but we don’t know how much. It’s possible that I could create a rating system for top players based on the Five Factors when I’m done with them that would be able to work around all of the exclusivity problems of a seven-player game. Maybe before I die I’ll do that.

SO WE REALLY DIDN’T LEARN ANYTHING FROM THIS ONE?

Rude! Here’s what we learned:

  • Explosiveness Defense matters. If you want to make Royale or win, you need to lose big less often. You won’t win the league with just Explosiveness Defense, but it is difficult to get to Royale without it. This may be a warning to boom-bust types.  
  • League winners do prevent ‘explosive’ scores at similar rates. How or why or how much it matters is very TBD.
  • Defense may need to be viewed holistically to really make sense. I’ll check back on this idea as we move through the Five Factors.

SEEMS PRETTY OBVIOUS.

Agreed.

HONESTLY, I’M PROUD YOU HAVEN’T MADE AN “E.D.” JOKE ABOUT EXPLOSIVENESS DEFENSE YET.

I know, right! Plus we just have to share the best in class defenses then the article is over.

BY THAT, YOU MEAN THE WEASELS WHO WON’T LET YOU GET IT UP?

Booooooooooo. Anyways, here are the best Explosive Defenses among our toppies.

  • FOGEL 2017. 5 explosive games allowed in 21 games (23.8%)
  • TROTTA 2017. 7 explosive games allowed in 18 games (38.9%)
  • O’KELLEY 2012. 6 explosive games allowed in 14 games (42.9%)

SO THE BTOAB IS ACTUALLY THE BIGGEST THREAT TO HAVING AN EXPLOSIVE GAME ON ANY BOARD?

Gotta love it when the data fits your narrative!

I MEAN… YOU DID FINISH SECOND TO HIM IN YET ANOTHER THING.

I’m done with you and your dumb jokes.

 

 

{jcomments on}

Read 468 times Last modified on Saturday, 24 February 2018 14:02

Comments   

+1 # Bryan Pravel 2018-01-06 07:13
Love this series Jake. What is striking to me about this factor is that I suspect you can identify players who throw dots by evaluating their explosiveness defense rating (my hypothesis is that it will be high, which of course is bad). That’s not to say that there are not other factors in which dot throwing could still be a useful tool (like negotiations), but it does at least imply you want to be selective about who you are throwing dots to. From a pure score standpoint this data seems to imply frequent dot throwing should make it more difficult to make it to the Royale.

Question for the masses: So if you aren’t throwing dots and someone scoring explosively on you is bad for your overall league score, does this data give insight into what to do when you know you aren’t going to have a good or great score?
# Chris Kelly 2018-01-13 14:08
Quote:
... does this data give insight into what to do when you know you aren’t going to have a good or great score?
Like Bryan, I think this is a fantastic topic, but the question above is well put. Citing numbers is okay, but it would have been great to "interview" Brandon about whether his anti-explosiveness percentage is the result of any intentional strategy/tactics (as opposed to just being a coincidence).
# Jake Trotta 2018-01-15 10:08
Quoting Chris Kelly:
but it would have been great to "interview" Brandon about whether his anti-explosiveness percentage is the result of any intentional strategy/tactics (as opposed to just being a coincidence).


Glad you brought this up, as that is actually the long term plan. The quant side of the Five Factors can provide benchmarks of what it means to be a good or elite player. Qual (individual player interviews, group discussion, etc) is great for figuring out how to achieve those benchmarks.

I will note- it is likely too large of a sample size to indicate that Brandon's Explosiveness Defense is a coincidence. Whether it is deliberate strategy or just the virtue of being a good player Brandon can better answer.
+1 # Pete McNamara 2018-01-18 20:45
I would like to see the stat restated. I think it important to take out board tops for each person. For example, Trotta tops a lot of boards, so naturally he's going to score well on ED. You can have a high beta player who might board top or allow a high score or a true defensive player, who if he/she does not board top can still prevent others from having a high score.

Make sense?
# Jake Trotta 2018-01-24 12:52
Quoting Pete McNamara:

Make sense?


Totally- this is something that I'd like to do, but isn't feasible until we get the database online and into a different tool. Unfortunately, these articles still require a decent amount of hand coding, which is prohibitive of using larger sample sets.

You have no rights to post comments

Who's Online

We have 2193 guests and no members online

User

To request an account, contact us at info@windycityweasels.org.

Search