How accurate are the end of season stats? I'm struggling to believe mine...

I've thought for the past few seasons that I get terrible luck in ranked when I play Jarvan 4.

This season, I went 1W 5L in six ranked games on him before giving him up.

However in the end of season stats, it says I played 58 games on him, with an 84.5% win rate - or 49/9, of which 1/5 was in ranked. That means I went 48/4 on Jarvan 4 in normal games.

Now, I know I've done better on him there, but I'm struggling to believe I did THAT good.

There's no way it includes Nexus Blitz - I went 29/20 on him there. I checked only summoner's rift: I went 6/1 in normals and 1/5 in ranked. That makes more sense - I didn't think I played him all that much this season.

Aside from Nexus Blitz, my match history (which I went to normally to verify) has no other games. I know for certain I've played PvP Twisted Treeline games on him (probably 7-10), but none are in my match history. I wager I've also played 2-4 bot games this year on Jarvan (yet they're also not in my match history for some reason). Other than that, I can't figure out the other 45 games (less ~10 so still about 35). Is it counting only the wins from Nexus Blitz?

Anyways, I can't figure it out at all, which makes me very confused. Can I trust any of the stats then? All of my "best damage" stats are useless because they're all on Jarvan and I can't believe anything on Jarvan when I actually went 7/6 on him but the summary says I went 49/9...

My Ahri stats seem consistent with my expectation, except that it says I played Ahri Jungle twice (and only played Jarvan jungle 3 times). Oh, and I played Jarvan mid 3 times too? I'm practically certain I played Jarvan mid exactly once and I only played Ahri jungle in Nexus blitz (and played Jarvan Jungle in Nexus blitz at least 40 times).

Just, uuh... can anyone explain?