All-Star Let's Discuss Worlds Scoring

[King]

And this is where I think the problem lies. You can either get killed or rewarded depending on where you are in the OOA. Results should never depend on a flipping bingo ball, and in the large divisions such as small coed and small senior semi-finals, this is exactly what happened. I don't like it, but I think the concept works much better for finals, but it is a fail for prelims and semis - you can't score team 3 effectively if you have 35 teams left to score.

That is my biggest issue.

Out of curiosity for the teams that competed in the at large division AND round 1 of the internationals what was the first team in order of competition to make it to the next round? I would be curious. Anyone want to figure that out?

So in order that they appeared was the team that competed 5th in small senior the first one to make round 2?

 

[King]

For real where is the performance order and lets see. This would be a fun experiment.

 

[King]

Welcome to Google Docs

This did the work all for us. Thanks Emily Now if we just had a line graph of scoring by hour. Who can make a line chart!

 

[King]

Ok so I took it into excel, did a little regression analysis on small senior. I will see if I can get something more solid to show you, but if you were go to near the end of the day as opposed to the beginning your average score would go up about @20 points. This is a score based on picking a bingo ball out of a lottery.

 

[cheermommaRN]

And this is where I think the problem lies. You can either get killed or rewarded depending on where you are in the OOA. Results should never depend on a flipping bingo ball, and in the large divisions such as small coed and small senior semi-finals, this is exactly what happened. I don't like it, but I think the concept works much better for finals, but it is a fail for prelims and semis - you can't score team 3 effectively if you have 35 teams left to score.

Especially when Justin is in charge of said bingo balls...
"It's okay the Obsession ball has not been stolen..." Still one of my favorite parts of Worlds...

 

[King]

Especially when Justin is in charge of said bingo balls...
"It's okay the Obsession ball has not been stolen..." Still one of my favorite parts of Worlds...

Bwahahaha. I think the selection of the bingo balls was absolutely truly random. And highly entertaining.

Here is the chart. The trend line tells you what the average score was of the teams randomly selected. This number should not go up if the selection was truly random unless there is judging bias as the day goes on. This basically means if you are NOT Champion Cheer (the early on spike in the data and a top finisher who did not get a paid bid most likely because of luck) then you would fair much better going later in the order because your average score would go up for the same routine.

 

[King]

This is fun. Ill do it for the next few divisions. Which one will have the biggest time advantage I wonder?


The Fierce Board App! || iPhone || Android || Upgrade Your Account!

 

[King]

Ok we see for small coed there is the same effect. It wasn't as beneficial in small coed to go later, but still worth about 10-ish points.

 

[King]

And confirmation this exists is when you make the dataset small enough judges are not as concerned with saving room. In medium coed there were so few teams competing for top 10 there was basically no advantage.

(am I the only one really enjoying this? my day job is inventory forecasting and regression analysis so I am geeking out over this a bit. )

 

[cheernerd5678]

ahhhh I already had a math class today and now King is posting math on the boards. Too much for one day. I mean, it's fascinating but... numbers.

 

[dawgshow]

And confirmation this exists is when you make the dataset small enough judges are not as concerned with saving room. In medium coed there were so few teams competing for top 10 there was basically no advantage.

(am I the only one really enjoying this? my day job is inventory forecasting and regression analysis so I am geeking out over this a bit. )

View attachment 138

I'm an economist by training but never by trade. I LOVE graphs so please carry on.


The Fierce Board App! || iPhone || Android || Upgrade Your Account!

 

[kristenthegreat]

I'm on my phone so I'll comment in detail later, but I find this extremely fascinating

The Fierce Board App to Rule Them All! || iPhone || Android || Upgrade Your Account!

 

[King]

I don't think any of the others are valid because the data sets are no longer random. In IOC 5 the partial and the paid bids come at the end. There is no way I can say that the judges biased their scores because they came later because the good teams, in fact, did come later. Now how much of that is based on skill and how much of that is based on judging bias by knowing the paid bids come later is probably impossible to tell with so little data.

But we can prove it exists with small senior and small coed, and get backup to this from medium coed because of the extremely small dataset.

 

[Dixie]

The 2 day comps reward consistency, but Worlds rewards excellence.

It's not dissimilar to a team doing well in the season but catching fire in the playoffs.

I like it because there's pressure every day.

I agree, but who is to say Day 2 excellence is worth more than Day 1? I've never understood comps that are 20/80 or this compete twice, but only one score counts idea. Why not add the 2 days together? Or use the teams highest days' score from the weekend? Don't know how this works for larger divisions, but LAG makes sense because all teams competed twice with no prelims or semis.


The Fierce Board App! || iPhone || Android

 

[quitthedrama]

View attachment 137 Ok we see for small coed there is the same effect. It wasn't as beneficial in small coed to go later, but still worth about 10-ish points.

I love that you spent the time to show factual evidence. It was definitely more beneficial to go later.....not taking away from a few of the teams that made finals due to this, but the finals score cut off was less than a point and it really sucks for the teams that got beat by the "bingo ball effect". Hoping things change next year.

Sent from my MB855 using Tapatalk 2