Take them with a pinch of salt though. They don't have even a quarter of the data you need to actually read into them. In the best case, they are incomplete, and in the worst, misleading.
A few examples:
- A player with an insane meters run stat often didn't do it through strong carrying, but rather because they ran kicks back. This is why Ben Morgan often makes for good stat reading: he's one of the few 8s that will regularly drop into the back 3 to make the carry.
- Sometimes, a player can miss a tackle or two and put in a better defensive performance than a player that makes 100% of tackles. 20/2 makes better reading for a flanker than 5/0 for example. Also, players that blitz out of the line and miss the tackle can disrupt ball carriers enough so that a following team mate nails the next tackle, beyond the gain line. In this case, the speed out of the line of the first attempted tackle has to be noted.
- When comparing player against player, it's hard to compare number of tackles/lineouts won etc. For example, Gloucester made 2/3 of the tackles Northampton did after the first four rounds. If you compared Kvesic and Wood's tackle count, you'd have to bear in mind that Kvesic wouldn't be expected to make as many tackles as Wood.
- Often, teams with the better scrum "lose more" of their ball in the stats. This is because the team struggling in the scrum often aim to get the ball out asap and not give the team a chance for a turn over. They could "win" all of their put-ins with sloppy service for the backline. Comparatively, the stronger team aims to keep the ball in the scrum. Once or twice in the game, this may mean losing the ball. However, the longer the ball is in the scrum, the longer the dominating team have to push around the other team, and they can win good possession and penalties. So for example, team A may win/lose 10/2 of their put-ins whereas team B win/lose 11/0, but team A wins 6 penalties on their 10 successful put-ins, whereas team B win 0. This is why I think ESPN should have the scrum stats as won/lost/penalties won. The example would now read 10/2/6 and 11/0/0, and you could tell that team A was better in the scrum.
To put things in perspective:
Gloucester have won 80% of their scrums
Bath have won 78%
Leicester have won 76%
Saracens have won 74%
Northampton have won 73%