You're reading it everywhere. The San Jose Mercury News reports Democrats could win 6 seats in California. KPCC has an expert predicting that it could be 5-6 seats. This is pretty much the media meme. But what are the odds? I have my evaluation, but I'll save it, since it could have a bias.
I'm going to take a look at three highly respected forecasters, Charlie Cook, Larry Sabato, and Stu Rothenberg. They've rated each race. I put a numerical percentage to each. A district that's "Likely Democratic" will go to the Democrats 80% of the time. A district that's "Lean Democratic" will do so 60% of the time. One that's a toss-up will go to each party half the time. You can repeat on the GOP side.
Once we know each party's chances of winning a seat, we can put them into a probability matrix. We get this chart:
As the data in the bottom of the chart shows,if Democrats win all the Democratic leaning and toss-up seats they top out at D+4. That's a nice gain. But that's only if Democrats win every toss-up district. Toss-ups are classified as such because the forecaster thinks the district could go either way. That's the opinion the forecasters are given. Each party will win an equal amount. If the Democrats were to win all of them, then the districts weren't toss-ups after all. At the very least they should lean to the Democrats. If the Democrats are going to be likely to win 3 or more seats the forecasters would move the forecasts toward them. That's not happening.
It's a long shot that the Democrats will pick up more than 3 seats in California this year. Long shots sometimes come in, but the odds are better that Republicans pick up at least one seat. If the Democrats can get a long shot, then the GOP can do the same. The results this year should be something around Democrats picking up one seat. That's better than now, but not the bonanza they need to win back the House.