Tips on being ruthless/efficient with research?
This may sound bizarre for a lot of people here with IBD analyst background, but it’s the only venue I know worth asking… So here it goes.
I am doing a “pro-bono”/trial buy-side research for a hedge-fund friend, with not-so-hidden agenda of proving that I’m worth giving a shot as an analyst. I have breezed through building the model (got MBA in finance and CFA l.2) and recognizing the big picture, but now am stuck with the actual analytical part where thinking is required – coming up with defensible assumption ranges. The problem is abundance of online information, coupled with my perfectionism… There always seems to appear another angle of the issue worth investigating, which brings another 10-15 papers/reports, and so on…
For a sell-side work, it wouldn’t be much of an issue, I’d just settle on combination of reputable enough sources that collectively maximize valuation. But for buy-side I feel compelled to find the actual truth, and that creates this never-ending scope creep.
I am familiar with time/project management concepts and am keeping the project structured, with milestones and deadlines. But, in the absence of employer pressure (and the presence of full-time job) this research is moving so frustratingly slowly and makes me feel exhausted and guilty for still not having accomplished much. To preempt a possible explanation, it doesn’t mean that I’m unfit for buy-side analyst du-ty, it just means I’m doing something wrong… :-) (haven’t done this before at this level)
Any advice on how to deal with this? If you do buy-side research professionally, how do you draw a line of “good/credible enough” and move to the next piece? Is it some-thing that can only be learned during the 2 years of IBD analyst pressure cooker? Or is it the PM pressure that creates enough momentum to crush through indecision and confusion?
I realize everyone here is busy; any contribution will be appreciated.
You're not doing anything wrong. If anything, obsessing over the details and new possible angles of analysis is a good characteristic for a new analyst to have. Knowing where and how to draw conclusions, what "really matters" and what to disregard, and how to manage your time are things you can only develop through experience. And that's partly why the PM ultimately makes the call.
IBD? Not so much. The only way to get more efficient at analyzing investment opportunities is to analyze investment opportunities. You'll gradually get better at filtering signal from noise and learn from observation what factors ultimately result in price changes. Sometimes, that gut indecision you're feeling is indicative of there not being that compelling an opportunity/mispricing in the situation at hand, and the best course of action is to move on. But that sense also takes time to refine.
As an analyst, your job is to have a complete understanding of the facts at hand, all the checklist type of things - know the company, its industry, valuation, competition, return on capital, capital structure, etc. and be able to synthesize that information. From your post, it sounds like the main thing you need to do is think more probabilistically. Valuations are estimates, and there isn't always an objective truth that can be known in advance.
Instead, try to come up with your own best estimate of what you think a fair valuation would be in different states of the world. Filter information with the idea that it is relevant if it helps you do that. What do you think is the most likely scenario? What's the downside if you're wrong? How good could this be in a blue sky scenario? Try to make some reasonable assessments of what the investment is worth (relatedly, what the market price might do) in these situations. Your PM (friend in this case) will help you figure out if and why that combination of risk/reward might be an attractive investment proposition.
The rest of your research can be devoted to trying to figure out the likelihood of the states of the world you can envision - what risk factors would result in a downside case, what things lead you to your alternative hypothesis, etc. But first you have to describe them.
I've always been curious about this probability weighted stuff (bull/bear/base case scenarios), how do you come up with the probabilities? out of the clear blue sky? I mean it makes sense to me because you're giving yourself a margin of safety, but I feel like you'd have to have a standard 10 80 10 (bull base bear) deal but I'm not sure.
I am not a fan of explicit "subjective probabilities" that try to assign an arbitrary likelihood to events that don't have multiple trials. Sometimes people compute a share price based on expectation, but I think that is not usually a good idea. Just because you think something is likely to happen doesn't make it 80% likely, or 50 or 99. And a downside case is either plausible or it is not. You don't get 10 shots on goal with each situation, so you can't really say the bear case is 10% likely. For my part, I look for risk/reward tradeoffs that are skewed in my favor. Maybe that means a case where the downside, no matter how subjectively likely, is close to zero, but the upside is significant. Or maybe it's closer to equal risk/reward, but I'm very confident things will turn out the way I think they will. I size the latter much smaller than the former, for what it's worth.
Stock markets are not perfect prediction markets, and even if they were, buying something worth X in expectation at something less than X can still lose you a lot of money if the event doesn't happen. It can make sense in a very broadly diversified portfolio, but otherwise you're often talking about gambling. For example, you'll see sellside biotech analysts ahead of binary events like a PDUFA date ascribing probabilities to FDA approval based on the proportion of phase 3 candidates they've approved in the past or something stupid like that. In fact every situation is a little different and the quality of the data, management interactions with the FDA, and treatment landscape all play a role. Would you pay $75 for a drug in phase 3 that's worth $100 if approved and $0 if not? It's usually more complex than that, but sometimes that's what the decision boils down to.
In my opinion, the better biotech investors form an opinion, subjectively, of how likely approval might be, what the valuation is likely to be in each case (sure, price will fall if the drug fails, but given the rest of the pipeline and product portfolio, should it?), and invest only when they have limited risk of loss AND expect a positive outcome. But it's folly to try to put an exact percentage on it.
I've had to do exactly what you mentioned in order to smooth out my process, start a checklist. Using a single list for all different opportunities isn't a surefire way to bang out a pitch, but provides you milestones/reminders of the big pieces of the puzzle that you need to hit on. If you do it right, it becomes a bit easier to realize when there actually isn't much opportunity to squeeze out. I've seen friends reports, mine included, where they'll go through the whole analysis just to conclude that the security is fairly priced. That's fine since you'll have the analysis ready to go when it pricing gets out of whack, but is a waste of time if you're trying to prepare an immediate idea.
Give us more info? Equity, fixed income, or something else?
Large-cap? small-cap?
Do you have a narrative that you can summarize in less than 2 sentences on what the market is missing? If your thesis is simply that the sell-side's EPS projections are sandbagged, then you probably need to re-evaluate your angle.
In my opinion, you need to see something that nobody else sees coming. You need to go far beyond having a better DCF model. In fact, why do you even have a model?
I would echo the previous post. Ive been taught to have several operational/valuation outcomes, then probability-weighting them to determine the fair value.
Thanks a lot for the comments. Great point on probabilities vs singular truth! I didn't even think about this angle...
To add more background, it's US public equity, mid-cap, short play, has to do with commodities (not tradeable commodities but stock exposed to mineral commodities). I've built the model to see what drives the stock and what input levels it is currently priced for. It shows that the stock is priced for ongoing increase in product price well in excess of inflation, due to growing demand downstream and supposedly tight restrictions on supply expansion. I think these restrictions are being vastly overstated by the industry (at least the several public players in it), to make it look stronger than it is and support its overblown valuation, and therefore this price increase won't last as other entrants will ramp up supply. Not to mention the company's market share will dwindle as well... Not to mention a decent chance of downstream growth being overestimated too, in which case the whole market will shrink. My thesis has nothing to do with multiple compression, I think it's too unreliable and, for lack of a better word, shallow...
To clarify my issue (in addition to erroneously looking for one truth, as @tempaccount has pointed out), it is abundance of info from various sources. There are US Geological Survey, Energy Information Administration, API, then multiple state-level regulators and commissions, then industry groups, conferences and presentations, then company-level presentations... If I were working on a doctoral thesis of something, I'd love spend a few months and learn everything there is to learn on the subject, but you can't do that for every stock idea... Time is running out.
I realize a lot of it comes from experience, and it is supposed to be painful and frustrating, that's why it is not for everyone... Just hoped there are some "best practices"/rules of thumb etc. about coping with information overload. How do you guys stop yourself from reading more and more and say "f..k it, I've learned enough, let's move on" -- and remain confident that you've done sufficient homework?
Focus on identifying two or three catalysts that are really going to move the price. Sure, many things can happen but not everything is going to move the needle for a company. Identify those items that are not built into the stock price which essentially means if analysts/investing public are talking about it then for the most part it's built into the price. Ask yourself, what is it that everyone seems to be missing and what's the impact.
Don't think of yourself as a sell-side analyst. You can always pick out the sell-side analysts on conference calls. They'll ask questions such as "last year you had 53M in depreciation so do you expect 54-55M this year?". Those kinds of questions are irrelevant for the most part on the buy-side. Sift through the bull and think of three items.
Pretend you have to summarize everything on one page. What would you say and what are the points you want to get across?
Yes, targets are somewhat arbitrary but well thought out.
This has the makings of a really good short thesis - it's not uncommon to see downstream OEMs get growth multiples and a lot of excitement that disappears quickly when new capacity comes online and/or a commodity cycle turns. And when earnings miss, expect to see multiple compression too. They often go hand in hand. I'm thinking of the solar, biofuels, and lighting industries among others. One thing to watch out for is the degree of commoditization in the underlying product. Often a margin decline/competition thesis hinges on products being somewhat interchangable. If the company has a sufficiently specialized niche or long enough contracts all your industry modeling won't make much difference. Timing is always tricky, too. I would echo @finance_king that you should try to identify near term catalysts that could make the short work. If you would like to PM me the name I'll be happy to give you a more specific two cents.
Personally, I prefer not to stop reading more and more - learning a lot about random topics is half the fun for me. I try to keep turning over stones long after I've come to a conclusion and put the position on. Learning more can help build your conviction, but sometimes you discover something that changes your mind. Of course you reach a certain level where you're comfortable enough to invest, and then your priority becomes working on other ideas even as you may randomly think of new avenues of inquiry on existing positions.
Decision making under uncertainty is all about dealing with information overload. That's why a probabilistic framework may be helpful. But at the end of the day you have to pick out the handful of questions you think matter most to the stock price and answer them to your own degree of comfort.
It sounds like you already have everything in place, so just come up with a few scenarios for drivers and derive a range of valuation. Unless the company is actually committing fraud and you're able to lay out exactly what they're doing wrong, you won't be able to prove your thesis is 100% full-proof.
As long as you can answer somewhat confidently why...
"restriction are being vastly overstated by industry" "co's market share will decline" "downstream growth is overestimated"
and the valuation scenarios presents a favorable risk/reward situation, then you can take a stab at the short.
I understand you want to present a golden pitch to impress the pants off your bud, but investing is all about using the most simplest of assumptions to find opportunities that provide a r/r that fits your/your firm's profile.
whenever i'm lost on assumptions, i generally back solve to see what assumptions are required to make the thesis / company "work". If i can then get comfortable with what is need for it to work, then we're on to something.
Gentlemen, thank you very much for the input! It's very helpful and, at the risk of sounding overly sentimental, uplifting, to see such a support.
Wouldn't your first few pitches take a long time as there is no way you know what is noise that is too be filtered out. I mean, until you research the info, you'll never understand what it really is or how important it is. even after you analyze something, it's not always clear how it can be applied, so expecting to be efficient even starting out it's kind of opposite?
Just wanted to say...great thread so far. Tons of great info.
Seems to me newer analysts spend too much time and effort on the numbers and models (because this is more objective and discrete) and not enough on supporting a thesis. As others alluded to, this is the 'what are others (whether that's sell-side or the market) missing' or 'what can go right/wrong and when' type of stuff. And there are lots of investments that work for lots of different reasons, it doesn't have to be a silver bullet thesis to make money.
I think this is interesting because every job posting I responded to last year had a statement about "must have strong modeling skill" and every interview asked about modeling capability. But then when I actually was interning over the summer the model was mostly an after thought. I fell into the same trap of trying to make my model perfect like a sell-side analyst until I realized that every question I got was more about the story and "big picture" and the numbers the model spit out where only sort of glanced over.
Ut maiores laboriosam eos labore qui quidem. Ut fuga eveniet non suscipit molestias. Eum ea voluptates ex et. Non blanditiis sapiente hic saepe molestias maiores. Voluptate animi at et maiores voluptate. Et sit voluptas vel et.
Rerum repellat molestiae corporis omnis iste aut necessitatibus. Qui ex quisquam consequatur asperiores voluptatum adipisci.
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...