Was bored, decided to respond to some stuff here (as I have some experience in the area).
Errors is a massively underrated stat imo
Quite the opposite, at least in raw (or basic per game rate) form and the way you're thinking about them (more errors per game = bad). Errors correlate (positively, but not particularly strongly either) to ladder position – meaning that the more errors you make, the better your team is likely to be (one outlier here – the Storm – driven by Cronk’s astounding and never seen elsewhere statistical shape)… this is definitely a weak predictor though, and in causality terms it’s close to if a then b and c, rather than if b then c.
Minimal penalties against is the most important stat. Lower the better.
Then turnovers from errors.
Nope again, at least in raw or per game form. Similar to errors, so see above.
To scrape the surface of why this is (and it is really only scraping the surface), possession weights need to be applied, as well as understanding that many errors come from trying to create something. More errors will be made if you are creating more chances, and more errors will also be made if you have more plays with the ball in your possession – weighting error numbers for these factors will give a better indication.
Same for penalties, in a slightly different way (the best teams tend to get more into the tackle, with the result that they can lead tackle counts across a season despite having more of the ball – this is a weak correlation). More tackles (generated by more tacklers per opposition run) generally mean more scope for penalties. Also with the current trend for block plays, more attacking plays in opposition red zone will also mean more penalties against you. So again, penalty numbers need to be weighted – this time by tackle numbers and a possession/position consideration (to start with).
Wouldn't mind a stat recorded for:
1. Kicks caught on the full. Would cover their bomb defusing reliability and their general positional play;
2. Try stops - this might include cover tackles and try line hold ups etc;
Some of this is at least sort of seen through other stats that are publicly available (kick return metres are going to be higher for someone who consistently gets to the kick early, particularly when weighted against other things like linebreaks and tackle breaks). Try stops would be handy, they already have a 1v1 tackle count.. but being able to isolate events by field position helps this (1v1 in red zone is likely going to be a try saver).
Especially game-saving ones
Game-state (what the score was when an event occurred) weighted stuff has value, but more so when looking at specifics rather than general analysis. Position weighted stuff is going to be stronger, in a general sense (I don’t mean player position, I’m talking about literally where on the field a particular event occurs).
Stats in Australia are monetised. The general public only sees the most basic, while clubs etc pay for the good stuff. You'd be surprised the detailed level in while they are captured.
Stats everywhere are monetised (the US sports generally release more than others as a historical by-product of how their sports analysis industry came to be, but they have much more that is not publicly available). But yeah, the League stuff is still nothing close to what it should be for best value use.
I remember reading an article before the GF where the sharks couldn't beat the storm on stats. From memory, storm were ahead on everything in attack and defence except for offloads.
Whoever compiled that didn’t know that the hell they were doing (or were intentionally trying to arrive at a particular conclusion). The two teams were reasonably similar in many respects, the biggest differences were the Storm’s gang tackling leading to higher tackle/game numbers and a higher effective tackle rate (and their uniquely low error/penalty rate which is largely a function of how they play rather than amazing quality amongst their team) and a better kicking game. On the other hand, the Sharks’ forward pack was a much more varied monster to the Storm… loads more offloads and tackle breaks creating opportunities. The stats leading into the game were only one sided if you selected them to be, or if you didn’t know what you were doing.
Stats have some use, but i think some people over rate them massively.
I’d say it’s more likely that people don’t use them correctly (and stats – even in much better detail than currently in the NRL, and beyond what sports analysis can currently do in any sport – can never tell you the whole story, it’s important to recognise that - and specifically where those limitations are) than the stats themselves are overrated. 99.9% of people that look at stats don’t really understand what they’re telling you… your Burgess/Taylor example is an example of this, though I can’t be bothered pulling out precisely what is wrong with that comparison (at a quick guess without reference to stats I’d point out that Burgess probably played half or more of his minutes that year in the middle (13), not the edge (11-12), making the comparison immediately limited, if not completely invalid).
