Anyone trying to make sense of the stock-market plunge by focusing on President Trump's tariff yo-yo was left scratching their head at the start of this week. Shares in Tesla, run by Trump's chainsawer-in-chief Elon Musk, plummeted 15% on Monday to take them back below where they were before the election. Yet shares in General Motors and Ford, both much more exposed to tariffs on steel, Canada and Mexico than is Tesla, actually rose, bucking the wider selloff.

Tariffs were the wrong place to look for an explanation. Look deeper into the trading dynamics, and a picture forms of the broader market forces that sent stocks into correction territory this week.

The odd Tesla/Ford moves were more likely driven by what Wall Street types call capitulation. That's when traders have thrown in the towel and are closing out bets they had been holding on to in the hope of a turnaround.

Blame the army of individual day traders who obsessively follow Tesla (GM and Ford are both regarded with scorn by this crowd). These speculators finally gave up as Tesla lost all its postelection gains.

The irony for those who abandoned Tesla is that the stock then leapt on Tuesday and Wednesday (and Ford fell, while GM went sideways), before resuming its slide on Thursday. Trump peddling Teslas in front of the White House provided a justification for some to try to buy the dip. Also at play: another technical move driven by hedge funds.

Tesla was the single biggest short position of hedge funds at the end of January, the latest data available, according to Goldman Sachs. Hedgies' bet on Tesla's share price falling had been extremely painful when the stock rocketed higher after the election. The drop back made them some profits (or perhaps merely reduced their losses). That made it easier to buy back the stock to close the trade.

All this matters when trying to decide if the market is done with its correction. Stocks tend to overshoot reality both on the way up and on the way down, as investors get overexcited or depressed. After the election they overshot upward wildly, far beyond what was justified. If they overshoot down, it will be time to buy.

It doesn't quite feel like it's time to buy just yet. But here are three tests to help you decide:

Sentiment has already turned sour among private investors. Bears outnumber bulls in the weekly survey by the American Association of Individual Investors, and financial newsletters are more negative than positive about the outlook for stocks, according to the Investors Intelligence survey. Sentiment is a contrarian indicator, and when it's very negative it can be a good time to buy, as it becomes hard to get even more depressed. If the cloud of negativity lifts, it would help stocks popular with individuals, such as Tesla.

Among institutional investors, however, there's not enough sign of panic to make me want to go back in. Hedging in the options market has picked up, a sign of worry. But levels aren't even as high as when the Japan carry trade unwound suddenly last summer. I like to buy when fear drives stocks too low, but this is not that moment.

Leverage adds to the overshoots, as hedge funds and day traders borrow to buy stock, or have to sell in order to pay back their borrowing. At major market lows, leveraged traders are forced to close trades to repay debt. This accelerates the drop -- and helps find the bottom.

Monday gave a taste of the chaotic trading that results from the unwinding of leveraged trades, but hedge funds have barely started to cut their debt, according to the prime brokerage arm of Goldman, which lends to them. The small fall in leverage alone isn't enough to make me think that a rebound in stocks popular with hedge funds, such as Big Tech, is imminent.

Overshooting fundamentals is another sign we've hit a major low. David Kostin, chief U.S. equity strategist at Goldman Sachs, compares economically sensitive cyclical stocks with defensive stocks to try to work out what economic growth rate is priced into markets. He concluded this week that, after the bank lowered its forecast growth, stocks aren't pricing in a significant slowdown.

Goldman has plenty of company in cutting economic forecasts as weak data prompts a growth scare that has rippled through stocks. But economic forecasts are guesstimates, not science, and whether there's been an overshoot depends on what you think will happen to growth under Trump.

Investors who think the concern is overdone should think cyclical stocks, such as Ford and GM, are a bargain. The increasingly vocal group predicting recession will prefer defensives, or to get out of stocks entirely.

How to decide? The uncertainty created by Trump has clearly affected some household and CEO spending plans, but I'm unsure if the caution is widespread enough to tip the economy over the edge. This is hard to call, because it's impossible to know either what Trump will do next or how much the uncertainty itself will slow the economy.

Last week I wrote that I was paralyzed by all the uncertainty. The signs of at least some capitulation this week make me think a rebound is closer, but there isn't -- yet -- enough panic to trigger my contrarian buy-when-I'm-scared impulse.