National League predictions and analysis: what the model has scored this season

Our National League predictions come from an 8-factor conviction model that releases nothing until both clubs have played 6 league games. As at 27 August 2026 it has scored 72 National League fixtures this season and released none. The first rated board lands on Monday 31 August 2026.

How accurate are National League predictions?

Start with the published research, because it says very little about the fifth tier. The largest recent public accuracy study ran 12,337 matches across 26 leagues from September 2023 to March 2026 and recorded an overall 1X2 hit rate of 55.1%, against the 33.3% you would get from guessing. Not one of the 26 leagues it covered sits below a country's second tier, and the National League was not among them. The same study put the English Championship at 52.2% across 1,007 matches, in the bottom third of its own table, and found league hit rates running from 49.9% in Serie B to 64.0% in the Norwegian Eliteserien. Its conclusion was that predictability tracks how unequal a division is rather than how good the model is.

Two things follow for the National League. It is a competitively flat division, so it belongs in the hard half of that spread rather than the easy half. And almost nobody is publishing serious modelling work on it, which is the gap this page exists in.

Now our own number, with its limits attached rather than in a footnote. In the National League the model released 8 picks and all 8 landed, from 14 April to 25 April 2026, and 8 fixtures across 12 days is far too small a sample to prove anything on its own. On a Wilson 95% calculation that record sits inside an interval running from 67.6% to 100%, which is a formal way of saying the sample cannot yet tell a good model apart from a lucky fortnight. It is the smallest division sample we hold and the most fragile number we own.

Two details make it more useful than the headline. All 8 were the market favourite, at an average price of 1.48 and a range from 1.12 to 1.83, so this is not a claim to have found angles the market missed. And the highest conviction score on anything we have ever released is 8.2, on Carlisle against Brackley Town on 25 April 2026, which finished 3-0, and that too is a National League fixture.

What has the model scored in the National League this season?

Live from our own scoring engine on the morning of 27 August 2026.

The model has scored all 72 National League fixtures from 8 August to 5 September 2026 and released none of them. The average conviction score across those 72 is 4.27, the lowest of the four divisions we cover. Eleven of the 72 cleared the release threshold and were withheld anyway, and those eleven are most of the story: across all four divisions the model has scored 257 fixtures this season, 13 of which cleared the threshold, so the National League accounts for 11 of the 13.

The reason none of them went out is in the factor counts. Of the 72, 34 rest on a single factor, head-to-head record at full weight, and 7 carry nothing at all because the clubs have no recent meetings. Thirty-one have now reached two factors, which is progress on last week, when this page reported that nothing in the division held two. It is still the thinnest card of the four at an average of 1.33 factors, against 2.73 in the Championship. Across the 185 settled fixtures the model called in last season's run-in, the average fixture was built on 7.11 factors.

The division holding the least evidence produces the most confident-looking scores. With one factor at full weight there is nothing underneath it to disagree, so the number goes wherever that factor points.

That is what the four-game gate is protecting against. One of the 72 scored a perfect 10.0 as at 27 August 2026, Southend against Kidderminster Harriers on 28 August, which has since been played and which our own settled record marks as a wrong call, and that score rested on nothing more than the home side having won 2 of the last 2 meetings. The best score we have ever actually released is 8.2, so the biggest number the model has produced in its life is sitting on this division's card, unreleased.

The other one has already gone. Worthing against Boreham Wood on 31 August was a 10.0 on 20 August and scored 7.8 as at 27 August 2026, once a second factor arrived underneath the head-to-head and disagreed with it. It never reached a user, and that is the gate doing the only job it has.

What happened to the withheld high scores that have been played?

Five of the eleven National League fixtures that cleared the threshold have now been played, and two of them landed.

Two from five, 8 to 22 August 2026, and every one of the five rested on a single factor of the eight. None of those five scores was ever shown to a user, so none is part of our record. We publish them because they are the strongest evidence for the gate that we hold, and that has to include the weeks the withheld scores look good.

Compare the released record, where calls at 7.0 and above landed 19 times in 22 across 14 April to 2 May 2026, every one built on a near-complete factor set. Same threshold, same model, five fixtures against twenty-two, and the only material difference is the weight of evidence underneath the number.

Last week this page read nought from three and drew no conclusion from it. Two from five deserves the same treatment. Five fixtures cannot separate a working model from a coin in either direction, and a page that got sober about small samples only when the number was bad would be worth nothing.

The wider read settles it. All 36 National League fixtures played across the first three rounds were scored as home wins, and 17 finished as home wins. The three rounds split 17 home, 6 draw and 13 away. Round three alone produced 8 home wins in 12 after round two produced 4, and the model did not change in between. A model with one factor of eight has almost nothing to move it off home advantage, so what looks like form in its results is the division's home-win rate moving underneath a fixed call.

Why does the National League reach the threshold first?

National League clubs play twice a week through August, so they bank league games faster than the EFL divisions do. All 24 clubs in the division have 4 league games behind them by 28 August 2026. The Championship and League Two reach the same point in the first week of September, and League One later still.

That ordering is why this division matters more than its size suggests. It is the first place the model will have anything real to say this season, and it is the division where our record, small as it is, is best. The board exists to surface the handful of National League fixtures we think are worth backing, so on most weekends it will be short, and in August it is empty.

Where can you get National League predictions today?

Not from us, and the honest answer is worth more than the visit. Forebet publishes a free prediction on every National League fixture, including the 10.0 game we are withholding. PredictZ tips every game in the division, and WinDrawWin has published free tips across more than 140 leagues since 2003. If you want a number on a National League game this weekend, use them.

Two more places we lose. PredictZ and Forebet both cover National League North and South and we cover neither, so if your interest sits in the sixth tier this is not the tool for you. And our own eight-from-eight record, for all that it reads well, is eight fixtures. A division page built on eight results would be a thin thing to trust, which is exactly why the rest of this page is about what the model has held back rather than what it got right in April.

When does the first rated National League board land?

Monday 31 August 2026, the summer bank holiday, when 11 National League fixtures kick off. That is the first date on which every club in the division has 4 league games behind it, so it is the first complete rated board of the 2026/27 season in any division we cover. Championship, League One and League Two all follow on 5 September.

Until then, the pre-season read stands on last season's settled record and says so: National League 2026/27 season preview. The empty board across all four divisions is set out in why the board is empty, the gate itself in why we don't score fixtures until four games into the season, and the method behind the score in how the 8-factor conviction score works. The dashboard will show every National League fixture, all 8 factors and the working from 31 August.

National League current-card figures pulled live from Lower League Analyst's scoring engine on 27 August 2026: 72 fixtures, 8 August to 5 September 2026. Released-record figures from the settled National League set, 8 picks, 14 to 25 April 2026. Whole-set released record 19 from 22 picks, 14 April to 2 May 2026. Results verified against club and published match reports on 20 August 2026. Comparative accuracy figures from a public study of 12,337 matches across 26 leagues, September 2023 to March 2026, published 18 March 2026. Lower League Analyst is a research tool for people who bet on the lower leagues. It is not a bookmaker and does not take bets. No model can promise a result, and past performance is not a guide to future outcomes. 18+. Please gamble responsibly. begambleaware.org