If Connect Four is solved, what's the point if both players play perfectly?

[Gillodibilo]

I recently learned that Connect Four is a solved game, meaning that the player who goes first (Red) can force a win if they play perfectly.

That made me wonder: what actually happens if neither player makes a mistake?

If Red has a guaranteed winning strategy and both players know it, doesn't that mean Red will always win? And if that's the case, doesn't the game become less about strategy and more about who gets to play first?

[Ceaseless]

I recently learned that Connect Four is a solved game, meaning that the player who goes first (Red) can force a win if they play perfectly.

That made me wonder: what actually happens if neither player makes a mistake?

If Red has a guaranteed winning strategy and both players know it, doesn't that mean Red will always win? And if that's the case, doesn't the game become less about strategy and more about who gets to play first?

You can say that about just about any game, if both players play perfectly. If a specific player wins, it's about who gets that role. If it's a tie, then the game is always drawn. In games with luck, it's about who got luckiest. If games with unknown information, it's who guessed correctly, such as a rock paper scissors fashion.

That said, the "if both players play perfectly" is doing a lot of heavy lifting here. A game like tic tac toe is trivial enough. In Connect 4, at least for the base game, while it's been solved by computers, it doesn't guarantee a human's success. As you learn more about the game, you can play perfectly against easier positions, but with more nuanced defenses it gets harder and harder to find the win. As it happens, it can often be easier to steal the win going second than to find the guaranteed victory going first, for reasons that actually tie into why going first is stronger to begin with. Even strong players still lose. That's one of the reasons Connect 4 ratings have things like 700s. If players could regularly play perfectly no one could stand out enough to climb up there. I know I still make game changing mistakes, some I get away with because my opponent overlooked it just like I did, and others cost me.

[play4row]

worth noting the arena here is 9x9, not the 7x6 board that solve applies to. so even setting aside ceaseless's point about humans not playing perfectly, the board most people are actually rated on isn't the solved one.

[Ceaseless]

worth noting the arena here is 9x9, not the 7x6 board that solve applies to. so even setting aside ceaseless's point about humans not playing perfectly, the board most people are actually rated on isn't the solved one.

Do more people play Arena than regular? I'd have figured the base game was the most popular and observed in regular mode it was. So much so playing other formats got a bit tricky since it felt like more effort than it was worth to sort out the nuances to properly play it relative to how much I'd actually get the chance to do so. Is it just that all of the Arena board players pretty much only play Arena? It's not often I see a non-Arena format dominate regular play, while I see a convincing reason that Connect 4 could be one such exception, I have never looked into the activity of their Arena format in particular to see if that was actually the case.

I haven't played much 9x9, but even at a glance with Connect 4 principles there are some obvious shifts. The column height of 9 being odd means you can't maintain parity going up a full column like you can with the even 6, which also means the natural zugzwang parity issues for threats don't stay the same each column either, unlike the base game. The final column may still be odds to the first player and evens to the second like the base game, but if we did a second to last full column as well it would switch, whereas it's always the same for every column in the base game. Odd/even row threats should also switch how they affect the parity of a column, now odd row threats maintain parity while evens change it, the opposite of how they work with a column height of 6. Due to the previously mentioned parity shift for each odd row column, it also means the parity maintaining odd rows in 9x9 don't fall easily fall in line or all fail as easily their even counterparts in the base game.

[play4row]

no idea on the volume split honestly, i'd have guessed regular is bigger too. narrower point was just that arena runs the bigger boards, ian ta has it on 9x9 the last three seasons and 8x9 or 10x10 before that, so arena ratings aren't coming off the solved board.