Story of the Stats

[TheDMC]

NRL.com have a much better stats section this year, can do some solid analysis.

From a non-solid quick analysis, here are some that I find interesting compared to other years:

We have the third least offloads (usually we are top 3).
We have the fourth least missed tackles (usually we are top 5 in missed)
We have the fifth least errors (usually we are top 3)

We are right near the bottom in all run meters (has been the case also in recent years)
Near the top in line breaks (standard).

From a stats point of view you would think Johnson is a better player than Thurston..


Johnson 7 tries / 8 line brk / 11 try ast / 8 offld / 14 err / 12 line brk ass / 3400 kick m
Thurston 1 try / 1 line brk / 13 try ast / 4 offld / 5 err / 13 line brk ass / 4000 kick m

They have both run the same meters more or less.
Johnson does more tackling and misses fewer.
Thurston goal kicking quite a lot better.

Errors obviously an issue for SJ.

Clearly game management is not captured well in stats.

 

[ZEROMISSTACKLES]

How would you capture game management though? I'm not saying you can't, it just sounds like a lot of work, beyond just counting.

I thought Thurstons try assist would be more that just one over Johnson.

Interesting stats that clearly shows Johnson is working hard out there. Probably harder than Thurston who has Morgan to kick back on.

Johnsons error rate is high though, too high for a half. I think our whole team would be up there in the most errors department. Then again, it could be due to having doing too much and NOT have someone like Morgan on the other side. No excuses really, "your focus needs more focus"-Jackie Chan.

Another thing is consistency. Thurston is consistent. Nobody says "depends on which Thurston turns up" and that's why he's in line to become an immortal because he has and CONTINUES to bring it.

Johnsons up there with the best, it's just that some weeks he likes to vacate down south with the worst.

 

[JJ]

Take out the two Newcastle games and how does SJ look?

 

[TheDMC]

How would you capture game management though? I'm not saying you can't, it just sounds like a lot of work, beyond just counting.

I thought Thurstons try assist would be more that just one over Johnson.

Interesting stats that clearly shows Johnson is working hard out there. Probably harder than Thurston who has Morgan to kick back on.

Johnsons error rate is high though, too high for a half. I think our whole team would be up there in the most errors department. Then again, it could be due to having doing too much and NOT have someone like Morgan on the other side. No excuses really, "your focus needs more focus"-Jackie Chan.

Another thing is consistency. Thurston is consistent. Nobody says "depends on which Thurston turns up" and that's why he's in line to become an immortal because he has and CONTINUES to bring it.

Johnsons up there with the best, it's just that some weeks he likes to vacate down south with the worst.

Right on all counts except maybe saying Thurston doesn't work as hard. His effort is more on game management, kind of like Cameron Smith in that regard (despite different positions). Yeah hard to do stats on game management, but I'm sure the Americans would have it nailed if they were into league. Errors and going missing are Johnson's biggest issue. Plus game management skills...which is aligned with going missing I guess.

On the try assists - Johnson has been creaming them in his good games - but I would suggest quite different TAs than thurston -- SJ many TA on his skip to the outside and great cut out balls etc - ie off his individual running. Thurston more off two or three tackle build ups where he knows exactly where the defence weakness will be and then passes to his runners who also know where the gap will be. More structured in other words.

JJ yeah -- performance in losses or tough games should carry some weighting else flat track bully scenario. But to be fair, Thurston operates with a much better pack, a great halves partner, and all round top notch team. So in many ways much easier for him.

 

[Meth]

Stats aren't insignificant, but definitely don't tell the whole story (as ZEROMISSTACKLES alluded to). I think the issues with Johnson this year can't quite be captured with stats.

 

[JJ]

Right on all counts except maybe saying Thurston doesn't work as hard. His effort is more on game management, kind of like Cameron Smith in that regard (despite different positions). Yeah hard to do stats on game management, but I'm sure the Americans would have it nailed if they were into league. Errors and going missing are Johnson's biggest issue. Plus game management skills...which is aligned with going missing I guess.

On the try assists - Johnson has been creaming them in his good games - but I would suggest quite different TAs than thurston -- SJ many TA on his skip to the outside and great cut out balls etc - ie off his individual running. Thurston more off two or three tackle build ups where he knows exactly where the defence weakness will be and then passes to his runners who also know where the gap will be. More structured in other words.

JJ yeah -- performance in losses or tough games should carry some weighting else flat track bully scenario. But to be fair, Thurston operates with a much better pack, a great halves partner, and all round top notch team. So in many ways much easier for him.

SJ is pretty much a flat track bully imo...

But agree with the above, and Maloney summed Thurston up pretty well the other day, nothing he does is that exceptional, just he does it more often in games than anyone, and his work-rate/involvement is phenomenal... SJ drifts in and out, has carved Newcastle up twice (so?) - Thurston is like the energiser bunny, he keeps taking the ball to the line, probing and waiting for the opposition to screw up...

 

[KeepingTheFaith]

3rd least amount of missed tackles is interesting, although we leave so many yawning gaps at times it's not like the opposition have to worry about breaking tackles.

Try assists are such a difficult thing to measure properly. They only get credited to the person who puts in that last pass/kick. SJ only got 1 try assist against the Knights, but he pretty much set up every right hand attacking raid with his swerve/step/run. Ayshford and Lolohea on the other hand have 2 try assists each.

Would be very interesting to see what SJ and JT would do if they swapped forward packs. Some games were impossible for SJ to do anything (Storm and Raiders debacles come to mind). Consistency is something that needs to come from a whole team, not just one individual.

 

[RHYTHM AND STEALTH]

Stats are like mini skirts
They reveil a lot but dont show you the whole thing

 

[jaseg]

Stats aren't insignificant, but definitely don't tell the whole story (as ZEROMISSTACKLES alluded to). I think the issues with Johnson this year can't quite be captured with stats.

Really depends on how you're using them. Raw numbers rarely can tell you much of any real value in terms of analysis, but with some work you can take plenty out of them. Still far too many things that aren't recorded (or recorded correctly) though. I'd want kick placement and (at least 2D) trajectories.. for a start. A key pass (separated from 'assist') metric would be very handy, too. As would dummy/lead runs. Just scratching the surface really..

The Johnson-Thurston comparison is a good place to start. Both have similar creation rates (this is actually very debatable depending on who is recording the stats - Prozone & Champion do their own counts with slightly different criteria.. the db I tend to use - for accessibility reasons - has a slightly bigger gap between SJ & JT in try assists), but a quick look at standard deviations will tell you that Thurston has been more consistent, and Johnson much more streaky (not just compared to JT but even his own previous seasons). Thurston has also seemingly changed his game this year - relying almost entirely on his passing game to create, since his running game just doesn't seem to be there like it was previously (even up to last year).

SJ's error rate is high, but most of the best players have high error rates (relative to their positions - with the occasional freak.. I'm looking at you, Taumalolo/Cronk/Boyd/Whare/Foran/Parker/Graham).. it's a little too high nonetheless, but not the biggest worry when looking at a wider view... after all, anyone care to guess which current half has the highest career error rate? If you guessed Thurston, you'd be correct.

As for the taking out Newcastle games- metres, runs & tries stay the same, try assists go (slightly) down, linebreak assists go (slightly) up, linebreaks go up (by an even smaller amount), tackle count goes down (obviously), effective tackle %s go down, offloads go up, tackle breaks go down - the biggest visible changes are conversion numbers (obviously) and somewhat surprisingly, his error rate.. it drops significantly.

 

[Penrose Warrior]

NRL.com have a much better stats section this year, can do some solid analysis.

From a non-solid quick analysis, here are some that I find interesting compared to other years:

We have the third least offloads (usually we are top 3).
We have the fourth least missed tackles (usually we are top 5 in missed)
We have the fifth least errors (usually we are top 3)

We are right near the bottom in all run meters (has been the case also in recent years)
Near the top in line breaks (standard).

From a stats point of view you would think Johnson is a better player than Thurston..


Johnson 7 tries / 8 line brk / 11 try ast / 8 offld / 14 err / 12 line brk ass / 3400 kick m
Thurston 1 try / 1 line brk / 13 try ast / 4 offld / 5 err / 13 line brk ass / 4000 kick m

They have both run the same meters more or less.
Johnson does more tackling and misses fewer.
Thurston goal kicking quite a lot better.

Errors obviously an issue for SJ.

Clearly game management is not captured well in stats.

What was the old saying - stats are like bikinis, you learn a bit but the real deal is hidden further beneath?

Like with the fourth least missed tackles, that's all well and good but we're third worst in points against this season. I'd guess we'd be lower in line breaks conceded. We have the third least offloads, but are in the top 8 for points scored. Fifth least errors and near top in LBs, but convert that into near bottom metres gained.

And as JJ said, Shaun's stats are heavily boosted by Newcastle. If you compared their performances v top eight sides, it'd be chalk and cheese.

League stats are pretty amateur. Look at golf, where a well-renowned mathematician took their basic stats and turned them into tangible stats like strokes gained in driving, putting, from 100m in etc which players on tour are actually taking to the range with them and seeing results they weren't able to gain before. In League, a 12-13m hit-up is wasted energy if the last 2-3 are painstakingly leg-driven to allow the defensive line to back-track. A 10m landing on their chest with a quick play of the ball doesn't look as good in the stats but obviously is much preferable. I remember that being a criticism of Steve Price's game, that people felt he was too keen on stats and not quick play the balls.

 

[Dav0c]

Thurston is overrated. Always gets credit when it's not deserved. He just needs to throw a standard pass in the lead up to a try for the commentators to jizz in their pants.

Or maybe I'm just drunk and delusional.

Either way, Johnson is the better player, in an inferior team.

 

[Meth]

Thurston is overrated. Always gets credit when it's not deserved. He just needs to throw a standard pass in the lead up to a try for the commentators to jizz in their pants.

Or maybe I'm just drunk and delusional.

Either way, Johnson is the better player, in an inferior team.

errr no

Thurston is a much better player than Johnson- and I am not a hater.

 

[TheDMC]

Stats aren't insignificant.

Hmm, depends on the confidence interval Meth.

 

[TheDMC]

errr no

Thurston is a much better player than Johnson- and I am not a hater.

Yeah as much as I would love to say Johnson is best in universe, he isn't. JT is.

 

[Dav0c]

errr no

Thurston is a much better player than Johnson- and I am not a hater.

I know you're not. We will just have to agree to disagree. It's always going to be a matter of opinion rather than fact anyway.

 

[JJ]

Have another drink.

 

[ZEROMISSTACKLES]

Either way, Johnson is the better player, in an inferior team.

If I was 9 years old and only understood League as the way you play it in the backyard then Johnson destroys Thurston. However it's not backyard League to say the LEAST.

He's got a better step!..and faster!

 

[Dav0c]

If I was 9 years old and only understood League as the way you play it in the backyard then Johnson destroys Thurston. However it's not backyard League to say the LEAST.

He's got a better step!..and faster!

His stats are better than Thurstons despite playing behind an inferior forward pack. Prove me wrong.

 

[VillaTheGorilla]

I don't know why people compare him to Thurston....future immortal, been the best half comfortably since Lockyer retired...those kind of players come along very rarely.

They're different types of players, and while the likelyhood is that Johnson will never reach the heights Thurston has, he still has the chance to develop, and learn to steer the team around the park. Remember, Thurston didn't play rep footy until his 5th year in the NRL, took him a while to find his feet at this level.

 

[jaseg]

His stats are better than Thurstons despite playing behind an inferior forward pack. Prove me wrong.

Wait... really?

You're massively overplaying the 'bad team' effect - you want evidence? See 2009. Cowboys came 12th, not a good team... Thurston racked up 28 try assists that year. Johnson has never even got close to that in ANY season he has played (closest was 21 in 2013). The two seasons either side of that the Cowboys came 15th (so quite a bad team), Thurston played only 17 games each season and created 16 and 20 tries - Johnson has never created at that rate on a per game (or per minute) basis. So let's just put the team effect away.

As for Johnson's stats being better than Thurstons... well, no. His running stats are consistently better, as one would expect. But all up (including chances created and/or converted from running the ball), Thurston has out-created Johnson every year since Johnson's debut, on both a raw number and per 80 minute basis, with the exception of one - and that one is far from over (this year).

All that is essentially trying to put Johnson on par with JT by under-rating try assist plays (the hardest thing in the game to do consistently at a high volume). When looking at halfbacks, I rate an assist slightly higher than a try (and certainly a linebreak).. and when you isolate that particular category (not just try but linebreak assists too / separately) Thurston destroys him every year, including this one.

But here's the figures I have anyway, counting tries & linebreaks the same as assists, on a per 80 minute basis... let's do it year-by-year, every year since SJ's debut.

Thurston
11 3.63
12 3.08
13 2.90
14 3.71
15 3.48
16 2.69

Johnson
11 2.53
12 2.70
13 2.56
14 2.51
15 3.12
16 3.34

The only year Johnson > Thurston is this one - and that comes with a huge qualifier... the season isn't over yet. All this doesn't even take into account things such as the kicking game (Johnson's has been inconsistent at best even this season).

Of course, age is a factor as well - Thurston's legs may be going (I made a point about the significant change in his game in an earlier post).. but it's too early to tell. It could just as easily be a minor injury, a form/confidence issue, or pure random chance... sometimes random things happen and are sustained over a period of time (like the - for any fellow EPL fans here - multiple random things that Leicester rode to winning a league title), particularly over what is still a relatively speaking short sample.