Role of draws for Elo-Calculation

I think draws in chess should not be weightet as high as it is currently is.

Let me explain why I think this:

Example 1: Alot of decisive games with decent win/loss ratio

Example 2: Domination in win/loss ratio but alot of draws

In this example Player 1 absolutly dominates and wins 500 games and Player 2 only 7, but the Elo difference is less than the example above because of the draws.

If you would play like earlier that when you draw you play on until you have a decisive game, it would be way different. What do you guys think?

I mean a match is decided by the win/loss score, not by the draws after all. If for example Magnus Carlsen plays against solid 2600 GMs who happen to play very good and draw him, he loses alot of rating points without ever losing.

What is your guys opinion?

I think the calculation you show can't possibly calculate Elo rating since it makes no reference to opponent rating.

The Elo rating system makes sense to me. In Example 2, almost 97 percent of the games end in a draw, so I'd say that the two players in that case are more evenly matched. The higher the winning expectancy, the larger the difference in strength.

I think you're confusing yourself by thinking "500 wins is a lot."

But you have to remember this in context. He was only winning 1 out of 30. That's not domination. If we drew 29 games and I beat you on game 30 would you say I dominated you?

I'm reminded of a funny story of a friend who played a much lower rated player at his apartment... for 15 hours! I didn't know this of course, so the next day he says "Guess what, [player's name] beat me 10 games yesterday!" and of course I can't believe it, 10 games, so many... but he'd lost something like 100 so it didn't really matter.