What? They're both extremely unreliable, Graves can dash right past them (to then all-in you).
He dashes past you, which makes it very easy to throw the E slightly behind him. He can't avoid them without walking a long way around. You can then hit him from outside his range, and use W to get extra time to smack him with (this can be done even if he avoids the E because avoiding it requires him to move in a predictable pattern).
And that's just at early levels. Once you get up to level 5 and beyond, your autoattack range is such that you can bully him with long range autoattacks without having to expose yourself to the risk of a multiple-hit buckshot. Farming safely at this point is very easy, given that you can push quickly and from range. Bullying him isn't particularly hard either. You can take advantage of his short range in the same way a Caitlyn could.
You shouldn't lose lane often to Graves. Going even should be by far the most common result.
That is a part of playing Graves almost regardless of who you're against. His range is just too short. A good Graves shouldn't have a problem.
And a good Jinx shouldn't have a problem giving Graves a problem. This is just unrealistic. Range is going to be a problem for Graves, so champions who can more easily exploit the range gap are going to have a much easier time against him. The range gap may be mitigitable against an MF or a Corki, but it's going to be easily exploited by champions like Jinx. If the Jinx player is as good as the Graves player, she's going to have an easier time taking advantage of the range gap than he will closing the range gap, particularly since he has to get through a frontline to get to her while she doesn't have much pressure at all to hit him.
That's actually just ordered based off of number of games. If you click on the advanced statistics, you can see that Graves' win rate against Jinx has for some reason dropped 4.16% recently
Both of your claims here are incorrect. First, the order isn't based on number of games. If you look below, you'll notice that Caitlyn is ranked below Sivir, Draven and Ashe despite having more games than any of them (16219) followed by Varus who has very few games (1461) followed by Kalista (7858). The number of games is irrelevant to the ordering on that list. The ordering of the list is based on the change of the factors presented in the indepth stats.
Second, they also aren't saying that Graves' winrate against Jinx has dropped 4.16% recently. "Recently" has nothing to do with it. That stat is saying that Graves' winrate against Jinx is 4.16% lower than his average winrate. If his winrate is ordinarily 50.45%, then it is 46.29% against her. This is not in any way saying there has been a change in the matchup recently. In fact, I can guarantee you that their stats for this matchup have not changed significantly in the last 3 patches (that is how long I've been using the site, so that's how long I can speak for). You can see a similar relative winrate for the two champions on Lolking's buttom-lane matchup chart.
You should reevaluate this data. I'm really not trying to be impolite, and I'm sorry if it comes off like I am, but I don't think you fairly considered what they were presenting.
I'd say it's because they only considered 20,000 games in the analysis. To put that in perspective, Jinx gets played around 1.3 million times a month in NA.
20,000 data points is not an insufficient sample (particularly since the 1.3 million is all games, and not just games of this specific matchup), and they aren't showing any signs of unusual variance that would lead us to suspect it would be. Statistical significance can be easily achieved with those numbers. This is not a valid complaint.