drawdown management at MM pod
Say a PM at a MM is running $500mm with 5% drawdown limit ($25mm). I have been to a few interviews at tier-2 MMs where the biz dev / mgmt expect 7% to 10% annual returns ($35mm to $50mm). Now, a few questions for seasoned PMs :
1) What is the max loss a PM would sensibly size their book to (would it be around 2.5% = $12.5mm i.e. half of their limits). If this is the case then, on average, would it suffice to say that a 10% return is 4x Sharpe. Is that your experience of annualized returns or are the HFs fibbing?
2) For a long/short book, gross will be $1bn. What percent of that do PMs usually allocate to - is it around 80% ($800mm gross) - and then they can make 5% on gross ($40mm). What is the typical book sized to in terms of a max loss (is it still $12.5mm max drawdown)
3) what is the average recovery time and turnover for a macro book (e.g. running G10 STIR, FX strategies)
Thanks - appreciate responses from seasoned PMs.
Can't answer 2 as not in L/S equities so have never thought of the words gross or net.
1. Hard to size to a "max loss" in most asset classes, especially in RV strategies. Most think in volatility terms, e.g. I run a "20mm vol book" => 1.2mm daily stdev. As you draw down you either mechanically or behaviorally decrease that volatility (e.g. many firms directly cut your risk capital at intermediate stops). So on "500" with a 10% return target -> 5% annualized vol on 2 "expected sharpe" you'd target 25mm annualized vol for the book = ~1.5mm pnl s.d. per day. In practice most would likely run below that number in most markets unless "forced" by risk => Citadel. In a draw >2% = 10mm almost all PMs will choose to downsize risk proactively. In practice, a 10% return is very rare as most PMs do size smaller than their delivered sharpe ratio. 1-1.5 is generally good in macro/rates/fx. So a 5% return year roughly average, 7% good, 10% outstanding, assuming you're running 5% vol. Most run lower if they can because it's psychologically easier hence why 10% is a hard nominal return target to hit (very high sharpe if you're running lower vol).
3. Recovery time and turnover are very strategy and regime dependent. For example, STIR in SVB recovered very quickly, so far post-Iran has been still challenging in 2026. Turnover very PM-oriented. Some guys turn their book around extremely fast (especially things like duration leans), some prefer longer term structural expressions (e.g. almost all rates PMs ran forward steepeners for 6 months straight last year, or FX guys structurally short USD/long carry CCYs).
Thanks.
To point 1, assume you're running a 2 sharpe strategy. And the PM will cut proactively around 2% => let's just say rheir digestible max drawdown is 2.5%. So that would mean their peak topline return would be 5% (which from the biz dev's perspective is just 1x sharpe, given the PM supposedly had leeway all the way down to 5%). It would take a freakish year where the PM would generate 4 sharpe to come out with 10% returns. I think most PMs exaggerate returns when speaking with biz dev, but it seems like the biz dev also up the ante by expecting a 7 to 10% return stream, which seems like it also has to be discounted.
Another point is that if net returns of a tier 2 fund are 7% and they are on a pass thru model (call it 8% fees), then gross returns are 15%. Given these platforms are levered on average 3x, the average PM has to be making around 5%.
Thoughts welcome from seasoned PMs.
To your first paragraph, that's largely why many multis force you to increase risk or lose capital + why hiring is so consistently aggressive. Most funds struggle (especially in discretionary macro) to find PMs who take enough steady state risk, especially at the "emerging" 250-500mm thresholds. Behaviorally most BDs and business leaders view discretionary macro as a slugging pctage strategy and want to push their macro teams that are making money to upsize risk. Hence why every PM interview as a jr. to midlevel PM is focused around your capacity to scale (provided you have a well articulated strategy + some PnL track record). The worst case for BD in hiring is to hire someone on a 500mm book with a 5-10mm guarantee that comes in and doesn't take any risk. Also a reason that comp packages and book sizes are so right tail heavy (more than just previous pnl they want evidence of sufficient risk appetite).
What is also left out of most discussions is that while your stated drawdown is formulaic, what happens when you hit it does have SOME discretion. Glen will likely never get fired from MLP. There is internal support for PMs who have made money for the firm, its LPs, and its partners. A pod/team that has made money will likely get more support to run at vol target with the PM much less likely to be as concerned about a 2.5% drawdown. So success ends up being quite autocorrelated (which many PMs miss when they think about switching platforms for a bag). If you've made 9 figures for your employer they tend to be highly invested in your continued success and will look to generally work with you in tough times. If you're a new PM at a new firm they are much more likely to assume an early drawdown = bad/no strategy.
To your third paragraph, think leverage is a bit higher and you're also excluding unencumbered cash here. Funds run 60%-80% unencumbered cash that earns SOFR (usually + 10-30 from treasury desks). That cushions returns beyond your calculation for another 2-3% for investors and would bring your average down a smidge.
Thanks for the insights. Returning to running a much larger pod sleeve (think 3 yards, with $150mm drawdown limit), how do you typically structure your book in rates, fx.
1) how easy is it to transact 1mm idv01 n 10y - would you do that purely in tsy futures, swaps or a combination of the above two plus swaptions. Would 1 dealer clear that in one go and what would be the bid/ask you'd assume in bps.
2) what about max size in curve trades (2s10s, 5s30s etc).
3) Also max size in US vs GBP 2y spreads - how will you execute this (futures, swaps ?)
4) for FX, how easy is it to execute $5mm premium for a 3m expiry atmf options in cable or yen.
What I am trying to get to is that when you read about the big shots at tier-1 places running 5 yards etc, what size in dv01, curve, dxy exposure etc are they running in the portfolio, and more importantly what is the mechanism of accumulating such large exposures. What happens at unwind time in stressed conditions.
Appreciate answers from seasons PMs. Thanks
These are fairly easy questions to answer if you've traded these products.
1 - However you want to trade 1mm DV01 you can trade it without much on the exec cost front. Futures block, swap on SEF (tradeweb etc), screens in UXY/TY whatever. You would presumably choose the instrument based on what exposure you wanted (do you have a view on swap spreads; are you derivs only or do you have balance sheet/repo allocation?)
2 and 3 are determined by your risk limits. Max size going to be governed by scenario analysis/historical pnl vol and conviction, also the correlation with other strategies in the book. E.g. in 2026 almost all Europe RV trades like a combination of front end longs and 10s30s steepeners so if you have a bunch of correlated risk your outright curve/duration position may be limited but your proxy exposure is large. Again, implementation choice going to be based on PM and mandate. Most 3-5bn pods as you'd describe (there are very few of those in the world) would have the knowledge of what risk they want to take and a broad mandate/expertise across implementations. SFIM8 vs SFRM8, cash spread, or swap spread all valid implementations. A 3bn book should run ~10mm daily SD so can just take that and impute sizing caps based on historical vol/scenarios.
4. 5mm premium in 3m ATMF FX is not a large trade. That would be an easy trade to execute at once and would trade extremely tight (~.1v max)
Unwind-wise, they are certainly harder books to unwind than others. A big Gilt RV book is basically un-unwindable (ask Glenn Hadden), as are large vol RV or MBS books in times of stress. A book that predominantly trades futures in "systematic macro" probably a lot easier to unwind relative to the risk taken/pl generation potential. Funds will generally try to estimate a stress loss unwind cost of your book or strategy and will plan accordingly. Illiquid books tend to be penalized in the MM model vs liquid books. How you would accumulate a position similarly correlates to liquidity. E.g. if you want a 5mm dv01 long 10y USD swap spread position you're going to have to build into that in clips over days/weeks. If you want a 1mm RX/TY spread that can be done in a couple minutes. If you want to run short 100mm not'l USDJPY that's one click, if you want to run long 3mm of GBPJPY variance that will take a while.
Hello nerds
Hi sorry but I’m a bit confused by “drawdown” I know what it means but if you are up 200m and go to 150 then at millenium they will close your pod? Or is it only when you start to actually lose money?
You would almost certainly not be fired in that situation. You would potentially be asked to cut risk down depending on what your capital/target vol/var looks like.
yh always been curious what they mean
is it max peak to max trough
would it change if i my max peak was achieved say 2 days ago vs 2 months ago
If your stop-out is a $50m draw, it's extremely unlikely you're ever going to be up $200m to begin with.
yh 50m draw is 35m vol
at 200m PnL that is a sharpe of around 6 which is getting into HFT land……
Eius tempora et commodi. Iste autem odio quaerat.
See All Comments - 100% Free
WSO depends on everyone being able to pitch in when they know something. Unlock with your email and get bonus: 6 financial modeling lessons free ($199 value)
or Unlock with your social account...