Not sure if this came up when the 1YPEG was introduced, but Saul said a few posts back:
I’ve also noticed that I’ve evolved a little in my use of the 1YPEG. If a company has an unusually high rate of increase of earnings in the past year, I’ll sometimes look ahead to what it will look at at the end of the year.
Regardless of whether a company has “an unusually high rate of increased earnings in the past year,” why not use a 1 year forward looking PEG, or a fancy term like “1YFPEG,” for all stocks? While 5 years of future growth is rather unpredictable, the next year may be a little easier to predict or at least guestimate.
I understand the logic of looking at the past year of growth for the 1YPEG since it is a real number, but it still doesn’t actually predict future growth. If this board is going to gravitate to a popular new valuation metric, why not use something that is slightly forward looking? A 1YFPEG would give you a better idea at first glance than the 1YPEG as to whether a stock is cheap or not.
What do you think? Sorry if this was already discussed.
Jeff
P.S. Thank you, Saul and to all for this great board and all the insightful posts.
The problem with anything in the future is that it is a guess.
Yes, but the problem with anything in the past is that it does not predict the future. The future is what we care about isn’t it?
I do understand, we’d have numbers more all over the map than a single, concrete 1YPEG number, and we’d all have to use an asterisk to define the percent growth used for the calculation and perhaps another for where the predicted percent of growth came be it based on management’s prediction or a consensus estimate. I see the flaw, but if there were enough analysts covering a particular company, one could use their estimates as the standard number. I’d think it would still technically be more useful than a historical number, no?
Yes, but the problem with anything in the past is that it does not predict the future. The future is what we care about isn’t it?
Well, yes, that is the dilemma exactly. We know about the past and can measure it with presumed accuracy. We don’t know about the future and can only estimate it. But, it is the future we care about for investing. If it is a big old company who has a highly repeatable track record, then we can estimate the future with some accuracy, but we don’t really care because the future is likely to be like the past … baring surprises. Where we really care about the future is exactly when it is difficult to predict, i.e., when growth is rapid. That makes it very easy to make mistakes. And, even smallish mistakes can change the premise. And, for smaller companies, the consensus may not mean much since there aren’t many analysts involved and the analyst may be doing little more than parroting what the company says.
Of course, the real core point is that no one number should be the guide. The more numbers the better and the more understanding of the company behind those numbers better yet.
forward looking guess vs backward looking "kind of "facts
I say “kind of” because the books may have been cooked, or the results achieved like they were at HP, cheapening of the product, relying on past reputation to sell. It works for a few years until consumers figure it out. By then the CEO is retired. or the company could be smack in the middle of a disruptive innovation. Like Blackberry.
Most of the time analysts really don’t know what is going on. They may use complex spreadsheets but the data they put into them is mostly speculation. The CEO may not know what’s going on either. .
I remember Chambers saying everything was great at Cisco right before the rash.
If it was easy to get rich in the stock market everybody would do it.
The weather today will most likely be the same as the weather tomorrow
Yesterday’s weather is an Extreme Programming (XP) term to keep teams from over committing during sprints and iterations.
The story goes something like this. Once upon a time there was a government spent a boatload of money on a weather prediction satellite. It took years. Cost of millions of dollars. But they were finally able to launch a satellite that was able to accurately predict the weather about 70% of the time. Not bad.
It was then that someone realized if they simply said today’s weather will be the same as yesterday’s weather they’d also be accurate 70% of the time.
XP uses this concept to keep teams from over committing during sprints/iterations. It reminds us that a good predictor of the future is what we’ve done in the past.
We use ‘yesterday’s weather’ to compute the velocity of our agile project teams. Usually it takes 2-3 sprints before we can get an accurate predictor for the next sprints. Of course, we’re always incorporating new information and will continue to reflect on past results to recompute our velocity. Things could be causing us to trail down, or things could make us go up. The point being is we’re not going to overcommit to what we can do going forward if we let history be our guide.
Hope this helps draw a similar analogy to investing. I view the 1YPEG as a measure of velocity, which really resonated with me when I heard it. It is a technique we use for software projects all the time.
Yes, but the problem with anything in the past is that it does not predict the future.
Whoa, Jeff! What is THAT rather categorical statement based on? If you stop and think about it, you’ll be embarrassed that you wrote it. The past doesn’t ALWAYS predict the future, but it’s sure better than anything else we have!
Saul
PS The anecdote that someone just posted about the weather hits it. If you just predict the same weather for tomorrow that you had today, you’ll probably hit it 70% of the time.
Hope this helps draw a similar analogy to investing. I view the 1YPEG as a measure of velocity, which really resonated with me when I heard it. It is a technique we use for software projects all the time.
Also, look at EPS growth acceleration and you will get even more information.
Looking at the 1YPEG and tracking it quarter to quarter will be very helpful. If Revenues and earnings keep going up and the 1YPEG keeps coming down that would be the sweet spot. It could be a company that is under the Radar or a company that is always blowing out earnings.
The problem with anything in the future is that it is a guess
Yes, but the problem with anything in the past is that it does not predict the future. The future is what we care about isn’t it?,
I like the word Probability; which is a branch of mathematics that deals with calculating the likelihood of a given event’s occurrence, which is expressed by numbers.
IMHO the only way to pick ANYTHING is to know the past.
If we did not have probablity then no Vegas
If we do not have probability we have no bets on a triple crown winner.
Also:
No spouse (oh no)
No car, house, life or “hole in one” insurance period.
No Lloyds of London
No bets period.
No walking on the moon.
and “God forbid”, no stock market!
The Probability of something “happening or working” in the future IS based on what has happened or worked in the past. Thus, the importance of learning from the past as in the 1YPEG.
Jeff: Yes, but the problem with anything in the past is that it does not predict the future.
Saul: Whoa, Jeff! What is THAT rather categorical statement based on?
Perhaps it was too categorical.
As people on the board were gravitating to this new magical metric of the 1YPEG, I just thought it perhaps less useful than something that does in fact attempt to look to the future a bit. Not far, just a year.
I did like the weather analogy. But we still don’t just blindly assume last year’s growth of 100% for stock A, 150% for stock B, and 207% for stock C is going to be 100%, 150%, and 207% respectively this year as well, hoping for 70% accuracy! Though perhaps some do. As stated upstream, most are still looking for clues as to the “probability” of continued growth based on the velocity of this past growth. If we have reason to believe the velocity of the past growth will continue or we find no reason to think it won’t, then we can be more comfortable with a future 1YPEG or 1YFPEG. This is what most of us are already attempting to do anyway.
Post 9474: “I’ve also noticed that I’ve evolved a little in my use of the 1YPEG. If a company has an unusually high rate of increase of earnings in the past year, I’ll sometimes look ahead to what it will look at at the end of the year.” -Saul
Post 9511 on SWIR: “…which would be up 90% with a PE of 22.5 and a forward end of 2015 1YPEG of 0.25.” -Saul
However we all go about it and regardless of what we call it, I do enjoy partaking in this shared quest to “beat the market.” Thanks to all the responses.
Touching on the backtesting thread here, but if you could backtest looking at companies with low 1YPEG and accelerating EPS growth, with getting ‘out’ of the company when growth starts to decelerate, as in, you get months or a few quarters of warning signs, I wonder what the results will be.
Overall positive I’m sure. Obviously the problem with the pure numbers game is you’re not reading the 10-Ks and finding other non-numbers, but nevertheless potentially catastrophic, warning signs.
