Other forecasts gave Trump even lower chances than the 20 percent in Silver's forecast with numbers like those: five or ten percent.

True: in the period around the second presidential debate (9 October 2016), at least some prominent forecasters or models (e.g. NYT Upshot, Princeton Election Consortium, HuffPost, PredictWise, Daily Kos) gave Trump win probabilities of roughly 5–10% (allowing about 3–13%), and these were lower than FiveThirtyEight's. Mostly true: such forecasts existed, but the figures were slightly outside the range (for example 12–15%, or 1–3%). Mixed: only one marginal forecaster was that low, or the low figures came from a clearly different date. Mostly false: other forecasters were mostly at or above FiveThirtyEight's figure, with only isolated low outliers. False: no notable forecaster gave Trump below about 15% at the time.

Set before the research began.

1. E1 2016 President Forecast HuffPost · 3 October 2016

   “- Clinton 98.0% - Trump 1.7%”

2. E2 Uncertainty and symmetry in the US elections | Stats Chat Stats Chat (Thomas Lumley, University of Auckland) · 29 October 2016

   “at the time of writing, 20% vs 8% from the Upshot, 5% from Daily Kos, or 1% from Sam Wang at the Princeton Election Consortium.”

3. E3 2016 Presidential Election: The Day in Polls Newsweek · 28 October 2016

   “He has a nine percent chance of winning the election, the forecast says, compared to Clinton's 91 percent chance. On Thursday, Trump had an eight percent chance of winning.”

4. E4 Grading The 2016 Election Forecasts BuzzFeed News · 29 November 2016

   “- Daily Kos (8%) - The Huffington Post (1.7%)”

E2, dated 29 October 2016, shows several forecasters giving Trump lower chances than FiveThirtyEight's 20%: 8% from the Upshot, 5% from Daily Kos and 1% from the Princeton Election Consortium. That fits the claim's pattern, but it is nearly three weeks after the second debate. E3 (late October) and E4 (final pre-election figures) also come from later dates. E1 is dated 3 October, but its 1.7% matches HuffPost's final figure in E4, so it does not reliably show a value from around the debate. Because the low figures come from a clearly different date than the one the claim refers to, the criteria call for mixed.