Field Theory by Hadyn Green


On Averages

I'm writing this before I head down to watch the Munster game. I'm looking forward to it because I'm a big Munster fan. I've got the entire first season on DVD (ho ho, although seriously I do). But what I really want to talk about are statistics. [Note: the bit on the Munster game is below the pretty pictures]

We've talked here before about the commentators' use of the word "average", when what they usually mean is "bad". Now, it could be argued that they mean the global average and not the local average of the player in question. But I think it's safe to assume they don't.

So I thought I'd take a look at the statistics and see what exactly "average" was for a number of players. I used Tracey Nelson's stats as they are the easiest to find and are broken down nicely by player.

I selected six players in total: Ma'a Nonu, Dan Carter, Richie McCaw, Rodney So'oialo, Ali Williams and Jerome Kaino. The players were selected because they have been criticised or glorified for their performances (or I was just curious).

For each player I used the stats for: Tackles; Assists (tackle); First Three to the Breakdown; Missed tackles; and Missed Tackle Ratio (number of missed tackles divided by all solo tackles attempted). And I only analysed the games played so far this year (mainly because I hate data entry), so we've got data for 11 games (I don't have the numbers for the Samoa game or the Hong Kong game).

Here are the players averages in each stat for the games played so far this year:

TacklesAssistsFirst 3 to the BreakdownMissed TacklesMissed Tackle Ratio

Sorry about the awful html

From this it should be easy to see the Richie McCaw really is "Captain Tackles". In fact if it wasn't for the injury in Christchurch and not starting in Edinburgh his numbers would be much higher (we'll come to that soon). Also note that McCaw misses so few tackles compared to who many he attempts that his missed tackle ratio is effectively zero. So if you're an opponent best not to run at him.

It can also be seen that our tackles are generally one-on-one situations.

For each player I created a graph (click if you need the bigger version) showing the three main stats: tackles, missed tackles and first three to breakdown (first 3). And then for each of those data series I added in the linear derivation of that series (which can be taken to be the average performance in that statistic for that player).

nonu graph

For example we can see that Nonu is clearly increasing in his role at the breakdown while there is an overall slight decrease in his number of tackles and (thankfully) his number of missed tackles.

Here are the others

sooialo graph

kaino graph
mccaw graph

What is really cool is Richie McCaw's average for Tackles and First 3. The only games that dip below his average are the game where he was injured and the game where he was subbed on.

williams graph

The most interesting chart for me is Ali Williams. If you look at his First 3 stats you'll see that it oscillates wildly from game to game both home and away, against various opponents. He did get injured against England but still that's some crazy variation. To be honest I wasn't expecting that and thought I would see that pattern in Nonu's numbers.

When I compared the players to one another I used the First 3 stat because for me it's the only offensive and defensive stat. Which means it should show the least variation in a win or loss and so should be a better indicator of "work" (the averages should just different for backs, midfielders and forwards).

Let's look at Nonu and Carter first (who I called "backs").

backs graph

We can see that they started going in opposite directions, until Sydney where they became in sync (Edinburgh being an anomaly as Carter was a sub).

forwards graph

Now for the forwards we can see that McCaw is the beautiful and wonderful God of Tackling we all thought he was. Note that his linear average is higher than the overall maximum of any other player.

If I had more time I would do a comparison of more players (though there is are very few for whom there is a large amount of data) and analyse by differing variables (such as line breaks, kick recoveries, tackles broken, passes completed, passes caught etc) for differing positions.

The Munster Mash
NZ 18 - Munster 16
So I've watched the Munster game now (well most of it, stupid work) and what a horrible game that was. Even the fights were dull. Stephen Donald looked like a horse's arse most of the time, thank god he held on to it a bit more in the second half.

The haka was very cool; all four Munster men seemed to get really into it. But the earlier helicopter bringing in the match ball was a bit much.

We were munstered in the scrums (see what I did there?). Ben Franks looked out of his depth and was shown up every time (though the refs calls of "pushing" weren't very helpful). On the other side MacIntosh was complaining that his opposite was walking backwards. As soon as Afoa came in the scrum seemed to get better instantly.

Franks also wasn't so good with the ball in hand. While others seemed to be able to push over the first Irish tackler at will, Franks would fend and roll to the ground gaining very few metres.

I was informed, by my Irish workmate, that Munster #8, Dennis Leamy is a "feckin' ligind" in Limerick. Or so she was told numerous times while watching rugby there.

There were a few good passages of play, when the All Blacks had the ball in hand and our defence seemed very well trained. But could the Munster players stay uninjured for more than two phases? It was killing momentum in both directions. Stringer behind the rucks was slower than Andy Ellis.

The crowd was once again a sea of serenity during the kicks. We were joking that the shhhh-ing before the silence was actually meant to mimic the ocean, relaxing the kicker. It was also hilarious to hear the commentators whispering to each other, afraid of being shhhh-ed.

As I had to leave early I missed what was apparently the only good bit of work by New Zealand. But I'm just happy we didn't lose and that I'm not going to get bollocks from my Irish workmate for the next 30 years.

48 responses to this post

First ←Older Page 1 2 Newer→ Last

First ←Older Page 1 2 Newer→ Last

Post your response…

This topic is closed.