As we have been doing all week, today's Ramblings is keeper-league focused, looking at decisions fantasy managers have to make between which players are kept and which are thrown back into the player pool. We have a lot more questions today than normal so let's not waste any time. My last couple Ramblings covering other keeper questions, covering Dylan Holloway, Brandt Clarke, Quinton Byfield, Daniil Tarasov, and a whole bunch more, are here and here.

The first three questions revolve around goaltending, and we'll start with Ilya Samsonov:

How do you deal with an unsigned player like Ilya Samsonov? Of course it depends where he signs, but is the upside of a potential landing (via trade or signing) worth the roster spot for a competitive dynasty manager?

— EdInTheMiddle (@EdInTheMiddle) August 13, 2025

This is a tough question to answer in the middle of August because the real answer is, "it depends". Training camps are about five weeks away. Let's say that in four weeks' time, we hear that Frederik Andersen has torn his groin, resulting in missing six months, and Carolina signs Samsonov as the backup. Then, yes, he'd be worth a roster spot. Teams always need goalie depth, and that would be the best-case scenario.

There is also a scenario where he signs a two-way contract and only gets 10 NHL appearances for a team like San Jose or Pittsburgh. In that case he wouldn't be worth it.

I would hold off on this decision as long as possible, but unless there's a significant injury in the next two months to a goalie on what is at least a borderline playoff team, it's hard to see Samsonov having much value. For reference, the 2026-26 Dobber Hockey Fantasy Guide doesn't even have Samsonov among its 82 projected goalies on the draft list.

*

The next goalie question concerns Karel Vejmelka vs. Darcy Kuemper:

Vejmelka vs Kuemper
SV%, GAA, W, SO

— Your Pal Dan (@LTDan30728175) August 13, 2025

On the surface, this is an easy one and it's to keep Kuemper. I do think Los Angeles takes a step back this year thanks to replacing Vladislav Gavrikov and Jordan Spence with Cody Ceci and Brian Dumoulin, but this was still arguably the best defensive team in the league. If those changes make them 10% worse – which would be a massive decline – they'd still be a mid-pack defensive team. A starter on a defensively average team describes Vejmelka, so as for the team environment, Los Angeles' worst-case scenario is basically being the Utah Mammoth.

The case for Vejmelka is that he's 29 years old and signed for five more seasons, whereas Kuemper is 35 years old with two seasons left. So, if I had a keeper team that was rebuilding, I would probably keep Vejmelka, but if I were hoping to contend for a fantasy title in the next year or two, it's Kuemper.  

*

Our last goaltender question is for Stuart Skinner:

Would you keep skinner for a 10th round pick that will be around 150th-160th overall?

— Armin Safari (@ArminSafari1) August 13, 2025

This is one of those "yeah it's fine" deals. I am not big on Skinner in any way, and it might depend on how managers in your league value goalies, but he's probably a top-25 goalie off the board in most fantasy drafts. In two single-season drafts I've seen, he's gone around the 125^th pick, so keeping him in the 150-160 range seems fine.

*

Next up is a question that deals with both defencemen and goalies:

[KevinCKeeper]

As for the defencemen, the easy answers are Mikhail Sergachev and Moritz Seider. Brock Faber just is not a good offensive defenceman yet and Kirill Kaprizov's future in Minnesota is uncertain. Brandon Montour is 31 years old and on a bad team. Sergachev is 27 years old, is on a team on the upswing, and has averaged 59 points/82 games over the last three years. Seider, meanwhile, is 24 years old and has three straight 40-point seasons. With similar-ish salaries across the board, it's Sergachev and Seider.

In net is a tougher choice. Because this is a cap league, I have no problem letting Igor Shesterkin go. There isn't much daylight between New York's team defence and Montreal's, Montreal is on the way up while New York is on the way down, and Sam Montembeault is less than one-third of the cost while MacKenzie Blackwood is less than half the cost.

I would probably keep Blackwood just because Colorado is the better team between them and Montreal and will be better both this year and next. After those two years, Montembeault is due for another contract, and if he keeps up his play, he will be considerably more expensive than Blackwood is now. Even if Montreal surpasses Colorado in three years (a big 'if'), Montembeault could have a salary double that of Blackwood's.

*

Our next question digs deep with Joe Veleno, Nils Hoglander, and Alex Holtz:

Dynasty league, no drops, salary cap.

Scoring cats: G, A, SOG, Special Teams pts, Hits+Blocks, FOW.

What to expect from Joe Veleno, Nils Hoglander and Alexander Holtz?

Is Martin Fehervary's $6m extension worth keeping?

Is Tristan Jarry a sleeper bounce-back candidate?

— Ivan Demigod (@kavanyoung) August 13, 2025

I will get to Holtz first, and I say this as someone who has been a Holtz believer since his draft year: It is hard to see much upside right now. He is no higher than the 7^th winger on the Vegas roster and probably won't even have a regular role that leads to 70+ games.

As for Joe Veleno, he has the opportunity to work himself into a third-line role for Montreal, but he will start the season on the fourth line and may even be a health scratch at times (it looks to be between him and Sammy Blais). An excellent year from him would be 10 goals, 20 assists, 90 shots, and 130 hits.

The guy from this group is Nils Hoglander. At the very least, he is assured a regular role in the Vancouver lineup, which is not something Holtz can say, and maybe even Veleno. On top of that, here are his comparables over the last two years by goal- and expected-goal driving numbers at even strength:

Höglander has been excellent in consistent depth roles for his career, and is a guy who should start getting more opportunities. What he does with them is another matter, but 10 goals and 30 points is a reasonable baseline, and those would be very good seasons for Veleno or Holtz. Höglander has both the security and the upside, so he's the choice here.

*

The next question is a fun trio with Troy Terry vs. Aliaksei Protas vs. Dylan Cozens:

Troy Terry, Protas or Cozens – Rotisserie league with Pts – Goals – Shots – PPP – SHP4 – Hits + Blocks

— k-money.eth (@MoneyFromDaRose) August 13, 2025

Let's start with the projections from the 2025-26 Dobber Hockey Fantasy Guide:

• Terry: 21 goals, 33 assists, 178 shots, 28 blocks, 12 hits
• Cozens: 20 goals, 31 assists, 186 shots, 36 blocks, 195 hits
• Protas: 24 goals, 37 assists, 131 shots, 35 blocks, 34 hits

On the surface, it seems we can throw Terry back. The Anaheim power play is really holding crushing his value, so if that gets better, he can easily be a 25-goal, 65-point forward, but there isn't a lot of peripherals to go with it, and he doesn't kill penalties.

The thing with Protas is that I think he is a very, very good player but he's also coming off a season where Washington shot 14.3% with him on the ice at even strength, the highest mark for any forward with at least 1000 minutes played in a season since 2007. Recent seasons with sky-high team shooting percentages that fell off the year after include Robert Thomas and Michael Bunting from 2021-22 to 2022-23, Andrei Kuzmenko from 2022-23 to 2023-24, and Elias Pettersson from 2023-24 to 2024-25. Those four players dropped by an average of roughly 2.7%, so if Protas drops just to 12%, he probably loses 6-7 points. He also skated under 17 minutes per game and the only Washington forward to reach 18 minutes was Tom Wilson. There isn't a lot of TOI upside and he's not someone who will get much of a power-play role.

There are ice time concerns for Cozens as well because Ottawa has Tim Stutzle as the top centre and they like their checking line of Shane Pinto and Ridly Greig. At the same time, Cozens is locked into a top PP role that Protas does not have, and Cozens could very well post double the combined shots, hits, and blocks that Protas will.

With the top power-play role and massive edge in peripherals, my answer is Cozens.

*

I got an email about a keep-5:

Off the hop, we are keeping Darcy Kuemper, Igor Shesterkin, Mitch Marner, and Martin Necas. The only question is which defenceman to keep, and I would lean Josh Morrissey because of age. I wouldn't be shocked if Josi has one more great season in him but he's 35 years old and health concerns are starting to pop up. One more great season might be all we get from him.

I understand not wanting to let go of Elias Pettersson, but I continue to be concerned about his knee. Maybe it gets better, but maybe it doesn't, and it's enough to not make him a top-5 keeper from this roster.

*

Our last two questions come from BlueSky. The first deals with Owen Tippett and Joel Hofer:

If this is a multi-cat league, I probably keep Tippett. Even if he's just a 25-goal, 50-point guy, being able to produce the shots and hits he does is immensely valuable.

If it's not a multi-cat league, there is a reason to keep Hofer. It seems there is a desperation in net and if things are so dire that keeping Hofer is a consideration, then getting improvements in the draft might be a tough sell.

In a multi-cat format, I keep Tippett. In a points-only setup, I keep Hofer.

*

Lastly, we have a question that is impacted by picks in the draft:

It is probably a good idea to hold onto Matthew Knies. He isn't far off from being a player who can post a 30-goal, 30-assist, 200-shot, 200-hit season. That can't be let go.

What I will say is that I think there's an argument for keeping both Lukas Dostal and Sam Montembeault. It gives you a goaltending tandem you can keep for years, and you don't have to worry about having to take whichever goalie is available in the first round (probably someone like Jordan Binnington or Stuart Skinner), or missing out on a goalie run in the second/third rounds when you don't have a pick. With your tandem locked up, you can focus on nothing but skaters basically until the end of the draft, and you will be able to draft Knies, or a Knies-calibre player, in the first round anyway. There won't be many (or any) goalies with the short-term and long-term upside of both Montembeault and Dostál available with that first pick.

My official answer is Knies and Dostál, but I would also think long and hard about keeping two goalies.