Validating Darts Tournament Skill Levels: A Non-Intrusive Approach

[DaveC426913]

TL;DR Summary

Looking for a non-intrusive way of measuring skill levels in darts. Is this method valid?

Let me preface this by acknowledging there are probably a dozen ways to assess skill levels and run qualifying rounds for any kind of tournaments. I'm not inventing rocket surgery here.

But I have reasons for wanting to do it this way, for which I may have to explain my reasoning as you come up with alternatives to my approach.
1. The idea is to have fair teams. This is isn't a pole position like in Formula One racing. (The reasn I mention this is because it means the idea of qualifying rounds doesn't work. Highly skilled players are motivated to play poorly during qualifying rounds. they they get under-valued. And then come out fighting. Basically, the darts equivalent of "hustling".)
2. It needs to be very non-intrusive. I don't want to waste valuable playing time qualifying rounds, and I don't want people to feel put on the spot. I want it to be practcally invisible.

What I want to know is: is the following approach going to give me reasonably valid objective results (Not for this year, but moving forward for next year) ?

Here's my idea:

• I pick a dart night, this year, and give everyone a scrap of paper.
• On it, they write their name and their team.
• We are playing Cricket. So 3 20's, 3 19s 3 18's 3 17's 3 16's 3 15's and 3 bulleyes wins a game. In theory, the winning team's "points" should total 21 unless they do so well, they start scoring actual points.
• Every time they score any chalk, they mark a point on the paper. If they get a 17, they mark a point. If they get a triple 20, they mark 3 points. etc. (I don't care what or how they score, I only care whether or not they are getting chalk.)
• (They are motivated to play their best, since this is part of this year's game. Good players can't "game the system" by playing badly without throwing the game they are currently playing.)
• They do this for two games (or three, but I think three might try their patience).
• They hand the scraps in to me at the end of the night.
• Maybe I do this test again near the end of the season and average them.

I figure - with the proviso that everybody has good and bad nights - this will give a pretty rough indication of how good each person is (better than simply guessing - which is how we do it now).
Here's what I'm not entirely sure about, let me know if I'm missing something.

Say, tonight Team 1 is playing Team 2. They are both somewhat matched in skills. They all mark their individual scores on our scraps of paper.

Team 1 (21pts)
A IIIII IIII
B IIIII II
C III
D II
Team 2 (19pts)
E IIIII II
F IIIII I
G IIII
H II

And on the board next to them, Team 3 - the best team - is playing Team 4 - the worst team. They are unbalanced. They all mark their individual scores on their scraps of paper.

Team 3 (24pts)*
I IIIII IIIII IIIII
J IIIII I
K II
L I
Team 4 (11pts)
M IIIII
N III
O I
P II

* Team 3 did so well they scored extra points against Team 4.My questions are thus:

• Despite the fact that Team 4 got outstripped by quite a bit, are their scores still an objectively valid comparison when matched up with Team 1 and 2 scores?
• Or does the fact that Team 4 got their pants beaten skew their results in some way I cannot see?
• Does their losing so badly somehow stunt their scoring in the bigger picture?
• Once I remove them from their teams and chart the players' individual scores, are they still objectively comparable?

Ultimately, I want to go into next year's tournament with a player skill chart like this:

I: 15
A: 9
B,E: 7
F,J: 6
M: 5
G: 4
C,N: 3
D,H,K,P: 2
L,O: 1

Is this valid?

 

[mjc123]

The one thing that immediately occurs to me is that the better the winning team is, the faster they reach the winning score, so the less opportunity the losing team has to score. Team 4 might have scored more against 1 or 2 than against 4. Conversely, the winning team always gets the same number of points (with the occasional exception as above), so it's hard to differentiate between them. Likewise the balance of players, at least in a winning team; if J, K and L were better, I would not appear so good. Could you rank players/teams on "average score per n darts thrown"?

 

[DaveC426913]

The one thing that immediately occurs to me is that the better the winning team is, the faster they reach the winning score, so the less opportunity the losing team has to score. Team 4 might have scored more against 1 or 2 than against 4.

Ah. Good point.

 

[The Fez]

Yes, darts stats for games like Cricket are usually marks per round (MPR), so number of scores per 3 darts.

 

[DaveC426913]

Yes, darts stats for games like Cricket are usually marks per round (MPR), so number of scores per 3 darts.

Yeah, I could divide the total marks for a game by the number of rounds played in that game.

The downside is number of rounds is one more data point they have to track.