There is a better way to run presidential debates. Actually, there are several.

We are now four debates into the 2016 presidential campaign, and the emerging consensus is that the format stinks. Yes, there may be some entertaining moments, but nobody seems happy with the status quo.

Now that the Republican candidates are reevaluating how the debates should operate, I humbly propose four alternatives to the weird and generally hostile group interview process so in vogue right now:

1. Crisis simulations
2. Oxford-style issue debates
3. Candidates submit the questions
4. Inter-league play

Crisis simulations

As I've argued before, the skills of debating are not generally the skills we really want in a president. We want deliberate thinking, not quips and one-liners.

To get a better sense of how candidates would actually perform as president, we should put them through crisis simulations to test their decision-making ability under real pressure. See how they'd respond to a bank run or a terrorist attack.

As I wrote in a Washington Post op-ed before the first Republican debate:

"Great presidents make the right calls in moments of crisis. They reason through unexpected problems. They ask good questions and sort new information quickly. They cut through ambiguity and thoughtfully weigh trade-offs. A crisis simulation would test these qualities in ways the debate format obviously does not. We want leaders who can resolve a Cuban missile crisis, not those who will stumble confusedly through a Hurricane Katrina disaster."

"Instead of spending the week after a debate arguing over who had the best comeback or the worst gaffe, maybe we'd get the kind of Monday-morning quarterbacking that touches on some of the challenges of actually being president. Maybe voters would learn a few things about the mechanics of the executive office."

It's easy to make a bunch of promises about what you will do. It's harder to demonstrate that you have the temperament and judgment to be the leader of the free world.

Oxford-style issue debates

If it's debates we want, it's debates we should have. But what we have now are not actual debates.

For example, we could apply the Intelligence Squared Oxford-style debate format. Split up candidates into teams of two, depending on where they come down on various issues. Then let them debate an issue that divides them: Resolved: "We should voucherize Medicare"; Resolved: "We should have a 10 percent flat tax"; Resolved: "We should engage more aggressively in Syria."

Let the candidates actually debate an issue over the course of 30 to 45 minutes, getting into enough depth to meaningfully help voters actually understand it. Such a format would also clarify who actually understands the issues and who does not.

The Intelligence Squared format is nice because while it is moderated, it also allows debaters to ask some questions of one another. And it has a simple format for declaring a winner. The audience is pre-polled. The side that wins is the side that moves the audience in one direction or the other.

A network could make this into a weekly program for the next several months, cycling through the different candidates in various pairings.

Candidates submit the questions

If candidates don't like the questions they get from the press, how about they submit the questions themselves in advance? Then the moderator would ask the questions randomly. I'd like questions to be anonymous, so they'd be a little tougher.

In such a format, anybody could get any question. This would encourage candidates to ask questions they feel would help them to differentiate themselves, but also questions they might want to answer.

This would largely eliminate gotcha personal questions, since the random aspect would take personal questions out of the mix. Debaters could still submit questions about particular candidates' policies, though.

One obvious benefit: Candidates couldn't complain as much about the questions, since they'd be the ones supplying them and the questions would be asked randomly. Even if questions do wind up being "softball" questions, as Vox's Matt Yglesias has rightly pointed out, easy questions are often harder to answer, because views tend to expect more cogent responses.

Inter-league play

Republican candidates keep promising that they could beat Hillary Clinton in a debate. Wouldn't it be both entertaining and elucidating to see Clinton or Bernie Sanders or even Martin O'Malley get onstage with a few of the Republican candidates and actually debate them? Especially in one of the two formats I've suggested above.

I can't imagine Clinton agreeing to do this, given how risk-averse her campaign has been. But I can imagine Sanders or O'Malley going for it.

Right now both parties are having separate conversations in public. Why not have one conversation? Such a debate would clarify the real differences between the two parties in a way these separate debates do not.

Let's experiment!

The networks' current debate format is a poor choice for educating voters on the issues and clarifying choices. It fails in its basic function of evaluating candidates on qualities most necessary for the actual job of being president. It's basically a weird group interview process with hostile interlocutors who never give candidates enough time to actually explain themselves.

Obviously, debate formats depend on candidates and networks being willing to try something new. The current format is pretty straightforward and familiar. It's not particularly taxing on either networks or candidates. But it's also not serving anybody particularly well. So let's experiment. There are many debates ahead.

This post is part of Polyarchy, an independent blog produced by the political reform program at New America, a Washington think tank devoted to developing new ideas and new voices. See more Polyarchy posts here.