These are based on a single study from 1996. The study was updated in 2005, unfortunately I can't find it. Here's info on it:
Here's a synopsis someone posted on bb.com:
Modification Of The Ergogenic Effects Of Creatine Loading By Caffeine: 1835 2:30 PM -2:45 PM
[F-15: Free Communication/Slide – Supplements and Athletes: FRIDAY, JUNE 3, 2005 2:00 PM - 3:30 PM ROOM: Jackson E]
Harris, Roger C.; Sale, Craig; Delves, Simon K.
Univ. College Chichester, Chichester, United Kingdom
(Sponsor: William J. Kraemer FACSM, FACSM)
Vandenberghe et al (1996) J Appl Physiol, 80: 452–457 reported that caffeine (Caf) blocks the ergogenic effect of creatine (Cr). In their study Caf was withdrawn >12 h prior to testing, meaning that Caf withdrawal and not Caf itself may have resulted in the reduced effectiveness of Cr.
To determine the effect of Caf ingestion and withdrawal on the ergogenic effect of Cr supplementation.
Ten healthy subjects participated in this double-blind cross-over design, incorporating two 9d supplementation periods separated by 6w. Pre-treatment testing (T1) was followed by 3d Cr-loading comprising 8 ? (0.5 g Cr.H2O kg-1 body weight).d-1. On day 5, the dose was reduced to 4 ? (0.5 g.kg-1 bw).d-1. From day 5 a single dose of 5mg.kg-1.bw of either Caf or placebo (P) was administered with the 2nd dose of Cr. On day 6, subjects were re-tested 2h after taking Caf or P (T2). Subjects continued supplementation until day 9 and were then tested 24h later (T3). Tests comprised maximal voluntary isokinetic knee contractions (MVC's) from 90? leg flexion to full leg extension performed as 3 series (S1, S2 and S3), each separated by 2 min rest. S1 comprised 3 ? 30, S2 4 ? 20 and S3 5 ? 10-repetitions of MVC's with 60s, 40s and 20s rest periods, respectively, between sets. Values are means (SE).
One subject withdrew from the study because of acute GI distress. Four other subjects reported GI distress when receiving 8 ? 5g Cr.H2O.d-1, which eased with P ingestion but not with Caf. Following Cr+P, mean peak torque (PT) differed between T1 and T2 by +8.94 (1.62)% in S1, +6.79 (2.09)% in S2 and +10.21 (5.68)% in S3 and between T1 and T3 by +9.77 (2.32)% in S1, +8.35 (1.5)% in S2 and + 11.68 (5.59)% in S3. Increases in PT from T1 were significant (P<0.05) in S1 and S2 at T2 and T3. These increases in PT were abolished (P>0.05) when Cr+Caf was administered, with PT differing from T1 in S1, S2 and S3 by -0.22 (1.85)%, -2.09 (2.73)% and +5.73 (8.64)% in T2 and by +0.70 (2.02)%, +3.75 (2.08)% and +14.60 (6.68)% in T3. Although PT was less affected by Caf at T3 (24h post) than at T2 (2h post) the difference was not significant. The difference in response to treatments was significant (P<0.05) for S1 in T2 only.
Whilst the results appear to support previous findings in that acute Caf administration may negate the effect of Cr, we consider that GI disturbance with Caf administration may have influenced the outcome of the study.
Here's an updated study from 2002 that claims the same thing:
Conclusion The researchers believe that the findings from this experiment offer indirect evidence that suggests that facilitation of muscle relaxation may be important to the ergogenic action of creatine supplementation as well as power production during sprint exercises.
However, for the athlete in training, the key finding is that sustained caffeine intake, over a three-day period, negates the benefits of creatine supplements.
You're probably right, but at the same time I don't care, because I'm taking it for recovery and not increased strength.
I'll probably switch to decaf without telling you, just so I don't have to concede defeat, though.