As the name implies, the Bump and Run Reversal (BARR) is a reversal pattern that forms after excessive speculation drives prices up too far, too fast. Developed by Thomas Bulkowski, the pattern was introduced in the June-97 issue of Technical Analysis of Stocks and Commodities and also included in his recently published book, the Encyclopedia of Chart Patterns.

The pattern was originally named the Bump and Run Formation, or BARF. Bulkowski decided that Wall Street was not ready for such an acronym and changed the name to Bump and Run Reversal. Bulkowski identified three main phases to the pattern: lead-in, bump and run. We will examine these phases and also look at volume and pattern validation.

1. Lead-in Phase: The first part of the pattern is a lead-in phase that can last 1 month or longer and forms the basis from which to draw the trendline. During this phase, prices advance in an orderly manner and there is no excess speculation. The trendline should be moderately steep. If it is too steep then the ensuing bump is unlikely to be significant enough. If the trendline is not steep enough, then the subsequent trendline break will occur too late. Bulkowski advises that an angle of 30 to 45 degrees is preferable. The size of the angle will depend on the scaling (semi-log or arithmetic) and the size of the chart. It is probably easier to judge the soundness of the trendline with a visual assessment.
2. Bump Phase: The bump forms with a sharp advance, and prices move further away from the lead-in trendline. Ideally, the angle of the trendline from the bump's advance should be about 50% greater than the angle of the trendline extending up from the lead-in phase. Roughly speaking, this would call for an angle between 45 and 60 degrees. If it is not possible to measure the angles, then a visual assessment will suffice.
3. Bump Validity: It is important that the bump represent a speculative advance that cannot be sustained for a long time. Bulkowski developed what he calls an "arbitrary" measuring technique to validate the level of speculation in the bump. The distance from the highest high of the bump to the lead-in trendline should be at least twice the distance from the highest high in the lead-in phase to the lead-in trendline. These distances can be measured by drawing a vertical line from the highest highs to the lead-in trendline. An example is provided below.
4. Bump rollover: After speculation dies down, prices begin to peak and a top forms. Sometimes a small double top or a series of descending peaks forms. Prices begin to decline towards the lead-in trendline and the right side of the bump forms.
5. Volume: As the stock advances during the lead-in phase, volume is usually average and sometimes low. When the speculative advance begins to form the left side of the bump, volume expands as the advance accelerates.
6. Run Phase: The run phase begins when the pattern breaks support from the lead-in trendline. Prices will sometimes hesitate or bounce off the trendline before breaking through. Once the break occurs, the run phase takes over and the decline continues.
7. Support turns resistance: After the trendline is broken, there is sometimes a retracement that tests the newfound resistance level. Potential support-turned-resistance levels can also be identified from the reaction lows within the bump.

The Bump and Run Reversal pattern can be applied to daily, weekly or monthly charts. As stated above, the pattern is designed to identify speculative advances that are unsustainable for a long period. Because prices rise very fast to form the left side of the bump, the subsequent decline can be just as ferocious.

Level Three Communications ( LVLT) formed a Bump and Run Reversal pattern after prices advanced in a speculative frenzy at the beginning of 2000. Prices advanced from 72 to 132 in 2 months and this advance ultimately proved unsustainable.

• The lead-in phase formed over a 3 month period from early Oct-99 to early Jan-00. Volume during this phase was relatively subdued and actually declined during the November and December advance.
• The trendline extending up from the lead-in phase lows formed a 34 degree angle. A visual assessment also reveals that this trendline is neither too steep nor too flat.
• The bump phase began in early January when the advance accelerated with a large increase in volume. A conservatively drawn trendline formed a 51 degree angle that was exactly 50% larger than the angle from the lead-in trendline.
• The distance from the lead-in phase's highest high to the trendline was 13. The distance from the Bump Phase's highest high to the trendline was 38. This is almost three times larger and validates the speculative excesses in the bump.
• After reaching a high around 132, prices declined sharply and bounced off the lead-in trendline. A lower high formed around 115 (red arrow) and the trendline was soon broken.
• The decline continued after the trendline break and reached 67 before a reaction rally began. The reaction rally advanced to around 95, but fell just short of the horizontal support line before falling back to new lows.