The Top 100's GW6 Bets: Isak Out, Newcastle In, Brighton Pays Off

The managers who finished GW5 in the world's top 100 have averaged 65.7 points in GW6 so far, after transfer hits. But only 43 of them remain in the latest official top 100.

Their biggest transfer move was selling Isak. Their most popular ordinary purchases were Gonzalo García, Josh King and Harvey Barnes. The first two have returned two points each; Barnes has not played yet. Meanwhile, Brighton's De Cuyper has delivered 15 points to a cohort that was already heavily invested in him.

So this week's story is partly about new bets, partly about keeping a successful core—and partly about captaincy choices that have gone wrong.

Snapshot: Sunday 11 October 2026, 19:25 BST. Nine of GW6's ten matches have finished, with confirmed bonus. Coventry v Newcastle is still to come on Monday. Scores include provisional automatic substitutions and can change before the gameweek closes.

Follow last week's leaders, then compare with today's

I pulled the 100-manager roster saved in FPL Cat's production database after GW5, at the GW6 deadline. I then fetched the latest official overall top 100 and the missing squads and transfer histories from FPL's public API.

The main analysis follows the original 100. That matters: looking only at today's leaders would select managers partly because their GW6 choices have already worked.

GW6 so far	Top 100 after GW5	Latest official top 100
Average points, after hits	65.7	84.6
Median points, after hits	67.5	86
Lowest–highest score	33–94	62–112
Average scoring players still to play	1.07	0.62
Managers taking a points hit	1	3
Bench Boosts	1	8

FPL's reported all-game average is 37 points. That is a useful reference, although it is an official gross average, while the cohort scores here subtract hits and include provisional substitutions.

The new top 100's higher average is a description of who has climbed the table, rather than evidence that those managers were a better group to follow before kickoff. Official ranks also have not yet incorporated every automatic substitution.

What did they buy and sell?

Among the original 100, seven used Wildcard, five used Free Hit and one used Bench Boost. Of the 88 managers outside Wildcard and Free Hit, 54 made transfers: 38 made one, 13 made two and three made three. Those managers made 73 transfers in total; 34 made none. Across the whole cohort, only one manager paid a four-point hit.

Here are their leading purchases. “Ordinary buyers” excludes this week's Wildcard and Free Hit managers; “all buyers” includes purchases recorded during those chips. Each count is managers buying the player, not the number of transfer transactions.

Player bought	Ordinary buyers	All buyers	GW6 points so far
Gonzalo García	11	20	2
Josh King	10	19	2
Harvey Barnes	8	14	Yet to play
Lewis Hall	5	15	Yet to play
Bukayo Saka	4	14	2
Kevin Schade	4	8	1
Leif Davis	3	8	1
De Cuyper	2	5	15
João Pedro	2	7	12

Isak was the clear exit: 15 ordinary managers sold him, or 20 including chip managers. Next came Szoboszlai with six ordinary sellers, Wirtz and Semenyo with four each, and João Pedro and Neco Williams with three each.

The most repeated ordinary transfer was Isak → Gonzalo García, made eight times. Gonzalo's two points beat Isak's zero, but this has not been a big immediate return. Josh King's ten buyers also received only a two-point player return.

The smaller Neco Williams → De Cuyper move, made twice, has worked much better: 15 points versus one. Both buyers scored De Cuyper's full 15 points. Two ordinary buyers of João Pedro also scored his full 12-point return.

Selling João Pedro went the other way. Three ordinary managers sold him, replacing him with Gonzalo García, Wissa or Welbeck. João Pedro has 12 points; those replacements have two, no match played yet, and one respectively. Those are player-point comparisons, not full team counterfactuals: price, bench placement and the rest of a transfer plan also matter.

The Newcastle bet is still open. Eight ordinary managers bought Barnes and five bought Hall. Their zero points currently mean “yet to play,” and Monday could materially change the verdict on the cohort's transfers.

Who did they captain?

Haaland was the most popular captain, but far from unanimous: 28 of the 100 chose him.

Captain	Managers	Player points	Captain points per manager
Haaland	28	6	12
Bruno Fernandes	16	1	2
Cole Palmer	14	8	16
Saka	13	2	4
João Pedro	8	12	24
Rogers	4	9	18

None of the original 100 used Triple Captain this week, so these are doubled returns.

Fernandes and Saka were the expensive captaincy misses. Together, 29 managers chose them. Fernandes captainers received ten fewer captain points than Haaland captainers; Saka captainers received eight fewer. João Pedro captainers received 12 more than Haaland captainers.

That is a substantial spread among managers who entered the week near the top of the same leaderboard.

The ownership bets that paid off

The table below compares all 15 squad places, including the bench. The original cohort's ownership is locked at the GW6 deadline. Overall ownership is the latest database reading and may include transfers already made for GW7.

Player	Top 100 after GW5	Latest top 100	All managers now	Original cohort's ownership gap	GW6 points
Haaland	87%	99%	74.7%	+12.3 pp	6
Calafiori	83%	95%	51.9%	+31.1 pp	9
João Pedro	82%	99%	68.0%	+14.0 pp	12
Groß	79%	90%	37.5%	+41.5 pp	6
De Cuyper	67%	91%	33.2%	+33.8 pp	15
Rogers	53%	76%	43.8%	+9.2 pp	9
Hall	52%	36%	20.9%	+31.1 pp	Yet to play
Verbruggen	38%	55%	21.5%	+16.5 pp	9
Barnes	32%	13%	8.2%	+23.8 pp	Yet to play

“pp” means percentage points. For example, De Cuyper's 67% ownership is 33.8 percentage points above 33.2% overall ownership.

Brighton has rewarded the existing template. De Cuyper returned 15 points, Verbruggen nine and Groß six. De Cuyper's ownership rises from 67% in the original cohort to 91% among the latest leaders. Calafiori and João Pedro have also delivered strong returns to widely held positions.

These were mostly existing holdings, rather than this week's fashionable purchases. Only two ordinary managers bought De Cuyper, and only one bought Groß. The main benefit came from already owning them and getting their points into the scoring lineup.

One example is Elgin City, ranked 63rd after GW5 and sixth in the latest official standings. They made no GW6 transfers, captained Rogers and have scored 94 points. De Cuyper supplied 15, João Pedro 12, Verbruggen and Calafiori nine each, and Rogers 18 with the armband.

That is the strongest score in the original cohort so far—and it came from holding the squad, rather than chasing a new purchase.

What Monday can change

The original 100 still have an average of 1.07 scoring players to play, compared with 0.62 among the latest top 100. Newcastle exposure is therefore one reason to keep this verdict provisional.

So far, the best results have come from the existing Brighton–Calafiori–João Pedro core, with captaincy making a large difference. The biggest new purchases have delivered modest returns or are still waiting for their match. Barnes and Hall are the unfinished part of the bet.

Follow the remaining match on FPL Cat's live points page, and compare your squad with the deadline cohort in the Elite Ownership dashboard.

Sources and method: production FPL database captured at 18:25 UTC on 11 October; live scoring data last refreshed at 17:25 UTC; official API collection completed at 18:28 UTC. Sources include FPL's Overall standings, per-entry picks and transfer histories, and stored official player scoring breakdowns. All 100 original managers have complete transfer histories. Recomputed player points match the stored GW6 totals; the ten differences from official current-cohort scores are fully explained by provisional automatic substitutions. Official ranks are used as returned by FPL, rather than inferred live ranks.

Reproducibility: scripts/top100-gameweek-snapshot.py and scripts/top100-gameweek-analysis.ts. Run from fpl-hub/: python3 scripts/top100-gameweek-snapshot.py --enrich --output data/research/top100/gw6-2026-10-11.json, then node --import tsx scripts/top100-gameweek-analysis.ts data/research/top100/gw6-2026-10-11.json docs/research/top100. The saved evidence and limitations are in docs/research/top100/README.md. Neither script writes to production.