Here are the biggest lessons that Bill Ackman learned in 2015 [emphasis mine]. And get on the free newsletter with insights not found elsewhere [subscribe here].

The first place to look for an explanation is mistakes we made in 2015, and we did make some important mistakes. Principally, we missed the opportunity to trim or sell outright certain positions that approached our estimate of intrinsic value. Our biggest valuation error was assigning too much value to the so-called “platform value” in certain of our holdings. We believe that “platform value” is real, but, as we have been painfully reminded, it is a much more ephemeral form of value than pharmaceutical products, operating businesses, real estate, or other assets as it depends on access to low-cost capital, uniquely talented members of management, and the pricing environment for transactions.

When we purchased Valeant at an average price of $196, we bought the company at a modest discount to intrinsic value as represented by the company’s existing portfolio of products and businesses, but at a very substantial discount to fair value in light of its acquisition track record,  the large number of potential targets, and its competitive advantages which include its low-cost operating model and favorable tax structure. When the stock price rose this summer to the mid- $200s per share, we did not sell as we believed it was probable the company would likely complete additional transactions that would meaningfully increase intrinsic value. In retrospect, this was a very costly mistake.

Our failure to sell stock wasn’t entirely an unforced error as we found ourselves largely restricted from trading during this period. During the summer, we were made aware of a large potential transaction that Valeant was working on, and as a result, we were restricted from trading at a time when it would have been prudent to take some money off the table. In retrospect, in light of Valeant’s leverage and the regulatory and political sensitivity of its underlying business, we should have avoided becoming restricted to preserve trading flexibility, or alternatively, we should have made a smaller initial investment in the company.

We made a similar error in not trimming our Canadian Pacific position when it reached ~C$240 per share. While we still believed CP was trading at a discount to intrinsic value at that price and there was the potential for CP to complete an industry-transforming, value-creating merger, in light of the size of the position as a percentage of the portfolio, and concerns we had about the Chinese economy, it would have been prudent to sell a portion of our investment.

Our most glaring, albeit small, unforced error was buying additional stock in Platform Specialty Products at $25 per share to assist the company in financing an acquisition. We paid too much as we assumed the new transaction would create substantial value, and because we assigned too much platform value to the company. Our assessment was incorrect as execution difficulties, operating issues, currency effects, and financing issues have destroyed rather than created value.

While not quite a lesson learned, as this has been a principle we have always believed, 2015 was also an important reminder that stocks can trade at any price in the short term. This is an important reminder as to why we generally do not use margin leverage in our investment strategy. We expect that there have been many margin loan liquidations in recent weeks which have contributed to dramatic stock price declines.

We do not believe that our investment performance in 2015 was primarily due to unforced errors, but rather due mostly to the market’s reappraisal of our holdings without a corresponding material diminution in their intrinsic value. While stocks can trade at any price in the short term, it is rare for companies to trade at material discounts to intrinsic value for extended periods. Fortunately, the lessons we have learned in 2015 should be easy to avoid in the future.