DIY Wallabies player ratings for Springboks Test: The results

Cracking article, Paddy!

He was the only one I gave a 4 to. Everyone else was at least a 5.

Pat, the trick is, do not communicate you are removing outliers and certainly don't talk about the quantum of data you will ignore. People (manipulators) will just vote more and add 2s and 3s to still lower the mean but even it all out. Ultimately I'd do a few more surveys, mention whether there were clear anomalies in votiong as you have done, and, if it continues, just quietly do as PeterK has suggested without notice.

Yep, I gave Hooper and Poey an 8 each for that reason Kaks. Hooper was everywhere in the first half and his workrate tremendous, but was more effective once Pocock was on and we were holding our own in the ruck and not going backwards. I gave 1 player a 4 and lots of 5's and 6's. To give out loads of 7s and 8's with a 9 or 10 in there as well I would want to see a match like that State of Origin 3 game by QLD where everything just clicked and everyone had very good games at worst.

Vee have vays of mayking you talk!

You should remove the outliers. People giving hooper 1's is just stupid. It does prove my own lifetime statistical analysis, and that is, 1% of all people are idiots.

just leave the data, hopefully there aren't too many petty people voting players they dislike at 1s. Trying to manipulate it no matter how good your intentions are will just bias the data further (unless you are trained in statistics).

I agree, basically you are manipulating the data. Although clearly there are outliers, I do not trust someone who obviously has something against players like hooper to exlude 10s as "outliers". What if someone played so well that the majority of scores are 8-10. Your bell curve method would be excluding scores of 10 which were deserved.

Willie 6 JPP 5 Kriel 7 Damian 7 Habana 7 Pollard 7 Pienaar 6 Beast 7 Bismarck 9 Jannie 7 Etzebeth 9 Lood 7 Louw 8 Coetzee 7 Burger 8 Bench 5 (without Lood, 4)

Think it works either way. We'd get better numbers if everyone used 5 as a pass mark but we still get a sensible result with everyone operating on their own scale. Average was 6.5, if everyone is a little happier next week, we'll see a higher number, less happy it will drop. Serves as a useful measure of the consensus view of where everyone sees each player's performance in the greater scheme. Really worthwhile new series.

I did give him the 10 for not making any mistakes. How do you say he played poorly if he didn't play? He fulfilled the role given to him by the coach perfectly - the bench under his backside remained warm!

Thanks Paddy, nice one. Pls continue! Maybe the one thing you might want to to add, is the link to stats page eg sanzars match stats.

But obviously if the Wallabies get flogged by 50 no body would get a 10 but the best performing player may get a 6 or a 5 & the worst performers will be in the 1-3s hope that makes sense?

Martin - I gave Hooper a 10 simply because IMHO I thought he was the Wallabies best on the night. It's ratings based only on the performance of the Wallabies players. Had it been ratings for both the Wallabies & Boks Hooper wouldn't have gotten a 10, I would of given it to Burger for he was the best player on the field by far IMO. As far as the ABs games or any other test that will have a player rating survey. Who ever I feel performs the best for the Wallabies in those games will get a 10 from me simple as that. 10 for me doesn't mean the player had a once in a lifetime game, it just means he was the best for the Wallabies in that game.

You would never give a 10 then which is wrong.

1- Worst game by a player in this position in Gold 2- Player was a detriment to his team 3- Player had a poor game 4- Below average game 5- Average game 6- Solid game, slightly above average 7- Good, but not great game 8- Great game, was a standout performer 9- Phenomenal 10- This man is a rugby God...Best performance by a Wallaby in his position

Please do. In basic terms we can identify which players are contributing most to the actual overall satisfaction fans give a teams performance, good or bad.

1. Beast (6.5) 2. BDP (8.5) This man should never have been taken off the field. He was incredible on the night 3. Jannie (6.5) if it weren't for his scrummaging, I'd have rated him lower. Didn't really impress me around the park 4. Etzebeth (8.5) Another outstanding performance from him 5. Matfield (Went off too early to score properly, but he looked to be going alright before the injury) 6. Louw (8) Man was creating Havoc on defense all over the place 7. Coetzee (8) This back row combination was great. When Vermulen comes back, it will be a tough decision which to drop.... 8. Burger (8.5) How many times did this man get over the ball to slow down Aus attack? Seemed like 100... 9. Pienaar (6) Rather ordinary game from him 10. Pollard (7.5) He showed why he should be the first choice 10 11. Habana (6) This is one of the very few games I have seen where Habana didn't really get himself involved a whole lot. 12. DeAllende (7) People were questioning his defense, I think he answered his critics 13. Kriel (7.5) Did SA just find their 13 of the future? Man he looked good on the weekend 14. JPP (5) Not bad, not good, not really involved 15. Le Roux (6) Quiet game from him, but he always looks dangerous with ball in hand

My Top scorer was Hooper, who I gave a 9 (I try not to give those, but honestly, the man had one heck of a game). I gave an 8 to only 1 starter, Folau, because he was so incredible under the high ball. Pocock, Holmes and Toomua got 8's for me as well, due to their impact off the bench completely changing the course of the game and. I don't think the Wallabies would have won without them coming on when they did.