One of the "truisms" in swaying is that timing is of minimal importance. Shettles' theories have largely been disproven, timing is listed as the least important of the seven swaying factors, and many assume that as long as the environment is right, timing really doesn't matter. Part of the reason for this dismissal of timing seems to come from the apparent lack of scientific studies supporting it.
One such scientific study that finds no evidence for timing is "Timing of Sexual Intercourse in Relation to Ovulation - Effects on the Probability of Conception, Survival of the Pregnancy, and Sex of the Baby." The study analyzes the last day intercourse occurred on and "[finds] no association between the sex of the baby and the timing of intercourse in relation to ovulation."
http://nejm.highwire.org/cgi/content/full/333/23/1517
The problem with this study is a problem I suspect a lot of timing studies have: it is looking at the wrong timing patterns. Most BD patterns do not sway, or if they sway, it is probably just by a small amount. There are only a handful of BD patterns I have found that may significantly influence gender (these are the patterns that I have at least 20 charts for and that have a greater than 55% effectiveness rate):
- 1 time BD the day before ovulation (23 charts): sways boy 65.2%
- 2 day cut-off (44 charts): sways girl 56.8%
- 3 day cut-off (22 charts): sways girl 68.2%
- frequent BD (at least 4 times) with a 1 day cut-off (26 charts): sways girl 69.2%
- frequent BD (at least 4 times) with a 1 day cut-off AND BD the day after ovulation (45 charts): sways girl 57.8%
* And, of course, I think O+12 also sways, but I can't evaluate it in my study.
If a study isn't looking for one of the handful of timing patterns that may sway, it is unlikely to find any evidence that timing works. For instance, the study above looked at the last day that intercourse occurred, and determined how many boys and girls were conceived. This is the graph of its findings:
There aren't numbers to indicate how many boys and girls were conceived with BD ending on each of these days; however, the article says 129 pregnancies resulted in a live birth, so there must be at least 129 pregnancies in the graph.
One of the obvious problems with looking at the last day of intercourse to evaluate timing is that BD patterns with a 1 day cut-off (for instance) can be very different. Both the best boy timing method and best girl timing method (in my opinion) have a 1 day cut-off. Yet the results are very different if there is no BD before the cut-off than if there is lots of BD beforehand.
Another problem is that the vast majority of all conceptions (at least by people who are trying to get pregnant) have intercourse the day before or day of ovulation. In this study, there were a total of 192 pregnancies initiated (including those that ended in m/c or stillbirth), and only 12 were from a 3 day or more cut-off (so only 6.25% of all pregnancies). This is similar to what I found in my study: only 38 charts out of 700 had a 3 day or more cut-off (5.4% of all pregnancies). I don't know how many charts had a 2 day cut-off in the study above, but the bars for the graph are significantly lower for 2 days before ovulation than they are for 1 day before ovulation. In my study, I only found 82 charts with a 2 day or more cut-off (11.7% of pregnancies). So, in other words, upwards of 88% of charts in my study had BD ending the day before or day of ovulation (and the percentage was probably similar in the study above). It is difficult to tell anything about timing by looking at the day intercourse ends, since almost all conceptions end on just two days.
A final problem is looking at too few pregnancies. Evaluating the sex ratio based on the last day of intercourse COULD identify whether cut-offs of 2, 3, or 4+ days influences gender. However, since cut-offs of 2 or more days are relatively uncommon, you have to look at a LOT of pregnancies to find very many cut-offs. Even with 700 charts, I still only have 82 cut-off charts in my study. 129 pregnacies is simply too few to be able to evaluate whether cut-offs sway.
As a test, I decided to look at the charts in my study to see if I also found timing of intercourse to have "no influence" if I only looked at the day ovulation ended. Since the study above looked at the last day of intercourse BEFORE ovulation, I left out charts that had BD AFTER ovulation. This is what I found:
- BD ends on O: 116 girl charts, 124 boy charts (sways boy 51.7%)
- BD ends on -1: 63 girl charts, 64 boy charts (sways boy 50.4%)
- BD ends on -2: 25 girl charts, 19 boy charts (sways girl 56.8%)
- BD ends on -3: 15 girl charts, 7 boy charts (sways girl 68.2%)
- BD ends on -4: 6 girl charts, 3 boy charts (sways girl 66.7%)
- BD ends on -5+: 2 girl charts, 5 boy charts (sways boy 71.4%)
If I include the charts with BD the day after ovulation, the results are similar.
- BD ONLY on the day after O: 1 girl chart, 0 boy charts (sways girl 100%)
- BD ends on O: 193 girl charts, 206 boy charts (sways boy 51.6%)
- BD ends on -1: 98 girl charts, 90 boy charts (sways girl 52.1%)
- BD ends on -2: 31 girl charts, 30 boy charts (sways girl 50.8%)
- BD ends on -3: 18 girl charts, 13 boy charts (sways girl 58.1%)
- BD ends on -4: 6 girl charts, 6 boy charts (50/50)
- BD ends on -5+: 3 girl charts, 5 boy charts (sways boy 62.5%)
In both analyses, the only significant result was that 2-3 day cut-offs sway for girl. All other patterns either had too few charts to be significant or had results close to what would be expected from chance. If I was just looking at the day intercourse ended, I would be able to tell that 2-3 day cut-offs swayed, but otherwise it would look like timing had no effect.
In conclusion, I do think timing is important, and I do think some BD patterns sway. However, since most BD patterns don't sway or sway very little, scientific studies are not going to recognize the importance of timing unless they are looking for the handful of BD patterns that influence gender. Simply looking at the last day of intercourse is not an effective way to evaluate whether or not timing sways.